Available online at www.sciencedirect.

com

Expert Systems
with Applications
Expert Systems with Applications 36 (2009) 894–908
www.elsevier.com/locate/eswa

An ANP-based technology network for identification

of core technologies: A case of telecommunication technologies
Hakyeon Lee, Chulhyun Kim, Hyunmyung Cho, Yongtae Park *
Department of Industrial Engineering, School of Engineering, Seoul National University, San 56-1, Shillim-Dong,
Kwanak-Gu, Seoul 151-742, Republic of Korea

Abstract

There have often been attempts to examine technological structure and linkage as a network. Network analysis has been mainly
employed with various centrality measures to identify core technologies in a technology network. None of the existing centrality mea-
sures, however, can successfully capture indirect relationships in a network. To address this limitation, this study proposes a novel
approach based on the analytic network process (ANP) to identification of core technologies in a technology network. Since the
ANP is capable of measuring the relative importance that captures all the indirect interactions in a network, the derived ‘‘limit centrality”
indicates the importance of a technology in terms of impacts on other technologies, taking all the direct and indirect influences into
account. The proposed approach is expected to allow technology planners to understand current technological trends and advances
by identifying core technologies based on limit centralities. Using patent citation data as proxy for interactions between technologies,
a case study on telecommunication technologies is presented to illustrate the proposed approach.
Ó 2007 Elsevier Ltd. All rights reserved.

Keywords: Analytic network process (ANP); Technology network; Core technology; Centrality; Patent citation

1. Introduction What is at the core of measuring technological interde-

Due to the intractable complexity and volatility of mod-
ern technologies, it becomes more important to photo-
graph the overall structure and internal linkage of
technology networks with the aim of grasping technologi-
cal trends and advances. Identifying and assessing techno-
logical advances critical to the company’s competitive
position is now recognized as a crucial activity for achiev-
ing and maintaining competitive positions in a rapidly
evolving environment (EIRMA, 2000). Since technology
systems are characterized by strong interdependence
(Archibugi & Pianta, 1996), there have often been attempts
to examine technological structure and linkage as a form of
network (Shin & Park, 2007; Wartburg, Teichert, & Rost,
2005; Yoon & Park, 2004).
*
Corresponding author. Tel.: +82 2 880 8358; fax: +82 2 878 3511.
E-mail address: parkyt@cybernet.snu.ac.kr (Y. Park).
pendence or linkage is patent information (Kim, Suh, &
Park, 2007). Patents have been the representative proxy
for technology (Trajtenberg, Henderson, & Jaffe, 1997).
A number of studies have been conducted to identify cur-
rent technology structure and make a projection of techno-
logical future trends by using patent analysis (Archibugi &
Pianta, 1996; Basberg, 1984; Basberg, 1987; Chen, Chang,
Huang, & Fu, 2005; Gangulli, 2004; Grupp, Lacasa, &
Schmoch, 2003). Several measures have been employed
for measuring technological linkage with patents, such as
co-classification (Breschi, Lissoni, & Maleraba, 1998;
Grupp, 1996), co-word (Courtial, Callon, & Sigogneau,
1993), and keyword vector similarity (Yoon & Park,
2004). Among those, citation analysis has been the most
popular one in spite of controversial discussions about its
validity. The underlying assumption is that there exists a
technological linkage between the two patents if a patent
cites another patent.

0957-4174/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.eswa.2007.10.026
 H. Lee et al. / Expert Systems with Applications 36 (2009) 894–908
Network analysis has often been used in conjunction
with patent citation analysis with the aim of grasping the
overall relationship and structure in a network. What is
at the center of interest is to identify important or core
technologies in a technology network (Shin & Park,
2007). As a quantitative measure of importance in a net-
work, centrality measures can be used in network analysis.
Among various measures, degree centrality has been
implicitly deployed as an indicator of importance of tech-
nologies in the previous studies (Trajtenberg et al., 1997).
However, it does not mirror indirect relationships despite
the fact that indirect citations as well as direct citations
play a crucial role in characterizing technology networks
(Wartburg et al., 2005). There are other centrality measures
that mirror indirect citations such as eigenvector centrality
(Bonacich, 1972) and reachability out-degree (Wartburg
et al., 2005). None of the measures, however, can success-
fully capture indirect relationships and produce meaningful
results for identifying core technologies in a patent cita-
tion-based technology network.
To address these limitations, this study proposes a novel
approach based on the analytic network process (ANP) to
identification of core technologies in a technology network.
Since the ANP is capable of measuring the relative impor-
tance of technologies that captures all the indirect interac-
tions in the technology network, the derived ‘‘limit
centrality” can be used as an implicative centrality measure
characterizing a technology network and showing core
technologies in the network.
The remainder of this paper is organized as follows. Sec-
tion 2 deals with the previous studies on patent analysis
and centrality measures in network analysis. The underly-
ing methodology of the proposed approach, the ANP, is
briefly introduced in Section 3. The proposed approach is
explained and illustrated with a case study in Section 4.
The paper ends with conclusions in Section 5.
2. Background
2.1. Patent analysis
Patents and patent statistics have long been used as tech-
nological indicators (Grilliches, 1990). Although patents
have been the representative proxy for technology as direct
output of R&D activities, there has been a ceaseless contro-
versy about the use of patent analysis since patents have
advantages and disadvantages like any other technological
indicator (Archibugi & Pianta, 1996). The pros and cons of
patent analysis are not explained here in detail, but can be
found in the literature by Grilliches (1990), Archibugi and
Pianta (1996), and Ernst (2003).
The most common method for early patent analysis
was to simply count patents and to compare how many
patents had been assigned to each entity, e.g. nations,
firms, and technological fields (Wartburg et al., 2005).
The basic idea is the more patents belong to different enti-
895
ties, the more important the entity is. Due to the highly
skewed distribution of patent values, however, judgments
on importance based on simple patent counts could be
biased to a large extent in many cases (Harhoff, Scherer,
& Vopel, 2003). It is also incapable of measuring
importance that mirrors influences or linkages among
entities.
Thus, what has become the center of interest in patent
analysis is citation information. Patent citation analysis is
based on the examination of citation links among different
patents (Narin, 1994). The use of citation information in
patent analysis boosts studies from various streams. One
of the main research topics is to measure the values of pat-
ents based on the number of citations of patents in subse-
quent patents. It is validated by a number of evidences that
more frequently cited patents have higher technological
and economic value (Breitzman & Thomas, 2002; Narin,
Noma, & Perry, 1987; Trajtenberg, 1990). In this context,
many studies have employed the number of citations as
an indicator of patent quality (Ernst, 2003; Hirschey &
Richardson, 2001; Lanjouw & Schankerman, 1999; Reit-
zig, 2004). Firm’s value can also be measured based on
the values of patents belonging to the firm (Hall, Jaffe, &
Trajtenberg, 2001). Another subject of studies with patent
citation information is to identify similarities between tech-
nologies. The similarity information can be used for identi-
fying technology overlaps with collaborative firms
(Mowery, Oxley, & Silverman, 1998), and proposing a
new classification system by clustering patents (Lai &
Wu, 2005).
The use of patent citation information in this study is in
line with the other research stream, analyzing technological
knowledge flows or technological linkages based on patent
citation relationships. However, patent citation analysis
alone cannot grasp the overall relationship and structure
among all the patents because it merely captures individual
links between two particular patents (Yoon & Park, 2004).
To address this limitation, network analysis, which will be
dealt with at the next section, has often been used in con-
junction with patent citation analysis to measure techno-
logical knowledge flows between entities and identify
important or core entities. A number of studies have been
conducted at various levels, such as national level (Jaffe &
Trajtenberg, 1998), industry level (Han & Park, 2006), firm
level (Ham, Linden, & Appleyard, 1998), and technology
class level (Shin & Park, 2007).
2.2. Network analysis and centrality measures
In general, the interactive relationships among actors
can be portrayed as a network composed of actors (nodes)
and interactions (edges) (Gelsing, 1992). The structure of
relations among actors and the location of actors in the
network provide rich information on diverse aspects of
an individual actor, a group of actors, and an overall
network (Marseden & Laumann, 1984). Thus, network
896 H. Lee et al. / Expert Systems with Applications 36 (2009) 894–908
analysis has attracted considerable interests from the social
and behavioral science community in recent decades, and
has also been applied and proved fruitful in a wide range
of disciplines (Wasserman & Faust, 1994). A patent cita-
tion-based network is one of the areas where network anal-
ysis is effectively employed with the aim of measuring
technological knowledge flows among actors. An actor
can be an individual patent or patents are assigned to a cor-
responding entity such as a nation or a technology class as
an actor. Then, the citation relationships among patents
represent interactions among actors.
To characterize either holistic network characteristics or
individual actor’s positions in a network, various centrality
measures can be calculated. Three common measures of
centrality are degree centrality, closenees centrality, and
betweenness centrality (Freeman, 1979). Among those,
degree centrality has been implicitly deployed as an indica-
tor of importance of technologies in the previous studies
(Trajtenberg et al., 1997). Degree centrality can be defined
as the number of ties incident upon a node. However, none
of these centrality measures take into account indirect rela-
tionships (Borgatti, 2005). Whereas in traditional network
theory indirect links are in general of less value than direct
links, this does not hold true in the case of patent citations
(Wartburg et al., 2005).
Eigenvector centrality is the one of the centrality mea-
sures that has to do with indirect influence. Eigenvector
centrality is defined as the principal eigenvector of the
adjacency matrix defining the network (Bonacich, 1972).
Simply put, the centrality of an actor is a function of
the centrality of actors who have relationships with the
actor; therefore, a node that has a high eigenvector is
one that is adjacent to nodes that are themselves high
scores (Bonacich & Lloyd, 2001). Although eigenvector
centrality has become one of the standard measures of
network centrality, it has rarely been employed where rela-
tions among actors have different strength or intensity,
that is, valued networks, since it only considers the cen-
trality of adjacent nodes and neglects how many nodes a
node is adjacent to and how much influence a node has
on adjacent nodes (Ruhnau, 2000). Also, influences on
actors that have no influence on any other actors are never
considered. Therefore, it cannot successfully capture indi-
rect relationships and produce meaningful results for tech-
nological knowledge flows in a patent citation-based
technology network.
Wartburg et al. (2005) proposed the reachability out-
degree to take into account indirect citations. The reach-
ability out-degree is defined as the probability weighted
direct Freeman out-degree times the probability weighted
direct Freeman out-degrees of the cited patents. While
the reachability out-degree can assess indirect citation
relationships, it is a proxy for specialization, not impor-
tance or impact. In addition, as what the reachability
out-degree opts for is unvalued networks where there is
no size in edges such as individual patent networks, it can-
not be implicative in the technology network where cita-
tion frequency determines the intensity of relationships.
In summary, it is required to develop a new centrality
measure that can capture indirect relationships in a
network.
3. Analytic network process
The ANP is a generalization of the AHP (Saaty, 1996).
The AHP, also developed by Saaty (1980), is one of the most
widely used multiple criteria decision making (MCDM)
methods. The AHP decomposes a problem into several lev-
els that make up a hierarchy in which each decision element
is supposed to be independent. The ANP extends the AHP
to problems with dependence and feedback. It allows for
more complex interrelationships among decision elements
by replacing a hierarchy in the AHP with a network (Meade
& Sarkis, 1999). Therefore, in recent years, there has been an
increase in the use of the ANP in a variety of problems such
as strategy selection (Wu & Lee, 2007; Yüksel & Dag devi-
ren, 2007), production-related decisions (Chung, Lee, &
Pearn, 2005; Lin, Chiu, & Tsai, 2007; Mulebeke & Zheng,
2006), project selection (Cheng & Li, 2005; Lee & Kim,
2000; Meade & Presley, 2002; Meade & Sarkis, 1999), logis-
tics decisions (Agarwal, Shankar, & Tiwari, 2006; Gencer &
Gürpinar, 2007; Jharkharia & Shankar, 2007; Meade & Sar-
kis, 1998), product design and development (Ayag& Özd-
emir, 2007; Gun̈gor̈, 2006; Kahraman, Ertay, &
Büyüközkan, 2006; Karsak, Sozer, & Alptekin, 2003; Wei
& Chang, 2007), product purchasing decision (Chang, Wu,
Lin, & Lin, 2007; Demirtas & Ustun, 2007), quality manage-
ment (Bayazit & Karpak, 2007), and financial forecasting
(Niemira & Saaty, 2004).
The process of the ANP is comprised of four major steps
(Chung et al., 2005; Meade & Sarkis, 1999; Saaty, 1996).
(1) Network model construction. The problem is decom-
posed into a network where nodes correspond to clus-
ters. The elements in a cluster may influence some or
all the elements of any other cluster. These relation-
ships are represented by arcs with directions. Also,
Fig. 1. Example of network in ANP and hierarchy in AHP.
 H. Lee et al. / Expert Systems with Applications 36 (2009) 894–908
the relationships among elements in the same cluster
can exist and be represented by a looped arc. Fig. 1
shows an example of the network model in the
ANP compared with a hierarchy in the AHP.
(2) Pairwise comparisons and priority vectors. Elements
of each cluster are compared pairwisely with respect
to their impacts on an element in the cluster. In
addition, pairwise comparisons are made for interde-
pendency among elements outside clusters. When
cluster weights are required to weight the superma-
trix at the next stage, clusters are also compared
pairwisely with respect to their impacts on each clus-
ter. The way of conducting pairwise comparison and
obtaining priority vectors is the same as in the AHP.
The relative importance values are determined with a
scale of 1–9, where a score of 1 indicates equal
importance between the two elements and 9 repre-
sents the extreme importance of one element com-
pared to the other one. A reciprocal value is
assigned to the inverse comparison; that is, aji =
1/aij, where aij denotes the importance of the ith ele-
ment compared to the jth element. Also, aii = 1 are
preserved in the pairwise comparison matrix. Then,
the eigenvector method is employed to obtain local
priority vectors for each pairwise comparison
matrix.
(3) Supermatrix formation and transformation. The local
priority vectors are entered into the appropriate col-
umns of a supermatrix, which is a partitioned matrix
where each segment represents a relationship between
two clusters. The supermatrix of a system of N clus-
ters is denoted as the following:
(4) Final priorities. When the supermatrix covers the
Ck is the kth cluster (k = 1, 2, . . ., N) which has nk
elements denoted as ek1, ek2, . . ., eknk. A matrix seg-
897
ment, Wij, represents a relationship between the ith
cluster and the jth cluster. Each column of Wij is a
local priority vector obtained from the corresponding
pairwise comparison, representing the importance of
the elements in the ith cluster on an element in the
jth cluster. When there is no relationship between
clusters, the corresponding matrix segment is a zero
matrix.
Then, the supermatrix is transformed into the
weighted supermatrix each of whose columns sums
to one. This ‘column stochastic’ feature of the
weighted supermatrix allows convergence to occur
in the limit supermatrix. A recommended approach
to obtaining the weighted supermatrix is to deter-
mine a cluster priority vector for each cluster, which
indicates relative importance of influences of other
clusters on each cluster. This can be done by conduct-
ing pairwise comparisons among clusters with respect
to the column cluster. The resulting priority vector is
then used to weight the matrix segments that fall in
the column under the given cluster. The first entry
of the vector is multiplied by all the elements in the
first matrix segment of that column, the second entry
by all the elements in the second segment of the col-
umn and so on. Repeating this weighting procedure
for all the column clusters produces the weighted
supermatrix.
Finally, the weighted supermatrix is transformed
into the limit supermatrix by raising itself to
powers. The reason for multiplying the weighted
supermatrix is because we wish to capture the trans-
mission of influence along all possible paths of the
supermatrix. The entries of the weighted superma-
trix represent only the direct influence of any ele-
ment on any other element, but an element can
influence a second element indirectly through its
influence on a third element that has the direct
influence on the second element. Such one-step
indirect influences are captured by squaring the
weighted supermatrix, and two-step indirect influ-
ences are obtained from the cubic power of the
matrix, and so on. Raising the weighted supermatrix
to the power 2k + 1, where k is an arbitrarily large
number, allow convergence of the matrix, which
means the row values converge to the same value
for each column of the matrix. The resulting matrix
is called the limit supermatrix, which yields limit
priorities capturing all the indirect influences of
each element on every other element. For more
details on supermatrix characteristics and theory,
see the text by Saaty (1996)
whole network, the finial priorities of elements are
found in the corresponding columns in the limit
supermatrix. If a supermatrix only includes compo-
nents interrelated, additional calculation should be
made.
898 H. Lee et al. / Expert Systems with Applications 36 (2009) 894–908
4. ANP-based technology network
4.1. Overview of proposed approach
The ANP underlies the novel approach to identification
of core technologies in a technology network. The ANP
and network analysis has the keyword, ‘network’, in com-
mon, but they are markedly different in ultimate objectives
and nodes that make up a network. The ANP is a MCDM
methodology aimed at setting priorities of alternatives or
selecting the best alternative. A network model in the
ANP is composed of decision elements such as goal, crite-
ria, and alternatives. On the other hand, the purpose of
network analysis is to grasp the overall structure of a net-
work consisting of a variety of types of actors by visualiza-
tion and quantification. When a network is constructed to
only visualize the overall relationships among actors, the
ANP has nothing to do with network analysis. If measur-
ing importance of actors or identifying core actors is
intended in network analysis, however, the ANP can also
be employed for the same purpose by viewing actors as
alternatives. Then, the centralities or importance of actors
are equivalent to the priorities of alternatives. That is the
basic idea of the proposed approach.
The overall process of the proposed approach is shown
in Fig. 2. Firstly, the scope and level of a technology net-
work is determined and patent data on selected technolo-
gies are collected. Then, a citation frequency matrix is
obtained based on the citation relationships among tech-
nologies. Finally, the ANP is applied to obtain importance
of technologies, which is named ‘‘limit centrality” that cap-
tures all the direct and indirect influence among technolo-
gies. In this section, the proposed approach is explained
with a case study on telecommunication technologies.
4.2. Technology selection and patent data collection
On any measure, the information and communication
technology (ICT) industry has been at the forefront of
industrial globalization (OECD, 2005). ICTs can be classi-
Fig. 2. Overall process of proposed approach.
fied into four categories: telecommunication, consumer
electronics, computers and office machinery, and other
ICT (Schapper, 2003). Among those, telecommunication
technologies have been playing a critical role in economic
growth and exhibiting dramatic technological progress.
Thus, analyzing the telecommunication technology net-
work is expected to provide valuable implications.
The primary source of patent data used in this study is
the United States Patent and Trademark Office (USPTO)
database. The USPTO has classified granted patents into
corresponding technology classes defined by the USPC
(Unites States Patent Classification). Each subject matter
division in the USPC includes a major component called
a class and a minor component called a subclass (USPTO,
2006). A class generally delineates one technology from
another and consists of subclasses that delineate processes,
structural features, and functional features of the subject
matter encompassed within the scope of a class. There also
exists a hierarchy among subclasses in a class. Every sub-
class has an indent level as a shorthand notation for illus-
trating dependency, represented as a series of zero or more
dots. A subclass having an indent level of zero is called a
mainline subclass which is set in capital letters and bold
font in a class schedule. Subclasses having one or more dots
are the child of a mainline subclass. As an example, a part
of the class 329 schedule is shown in Fig. 3. Among the 15
subclasses from 329/300 to 329/314, mainline classes are
329/300, 329/304, and 329/311.
In this study, a mainline subclass serves as a unit of the
technology network. The reason for using a mainline sub-
class, not a subclass is there is dependency among a main-
line subclass and the child subclasses. Since the child
subclasses inherit all the properties of their parent subclass,
it doest not make sense to treat all the subclasses at the
same level.
159 mainline subclasses in the USPC were selected as
telecommunication technologies. The IPC (International
Patent Classification) codes for the four ICT categories
are provided by OECD and shown in Appendix A. Refer-
ring to the US-to-IPC concordance provided by the
 H. Lee et al. / Expert Systems with Applications 36 (2009) 894–908 899

Fig. 3. Part of class 329 schedule.

USPTO website, the USPTO mainline subclasses matched Table 2

with the IPC codes of telecommunication technologies Form of citation frequency matrix
were only chosen. The selected 159 USPTO mainline sub- ... Class M (citing) ...
classes are shown in Appendix B and they cover 13 classes ... 1 2 ... m ...
in the USPC shown in Table 1. ... ... ... ... ... ... ... ...
Documents of all the granted patents assigned to the 13 1 ... fN1M1 fN1M2 ... fN1Mm ...
classes were collected from the USPTO database and Class N (cited) 2 ... fN2M1 fN2M2 ... fN2Mm ...
stored in our database. Since the number of patents is so ... ... ... ... ... ... ...
n ... fNnM1 fNnM2 ... fNnMm ...
huge that we cannot collect all of them in manual, the ... ... ... ... ... ... ... ...
own-developed JAVA-based web document parsing and
mining program was used for automatically downloading
patent documents. As a citation has a direction from a citing patent to a cited
patent, the matrix is asymmetric.
4.3. Citation frequency matrix To examine the current structure of the telecommunica-
tion technology network, citations made by patents only
The next step is to construct a citation frequency matrix granted from 2000 to 2004 were considered. The number
which represents citation relationships among mainline of citations for each cell was calculated by manipulating
subclasses. It indicates technological knowledge flows or the database storing the collected patent documents. Since
influences among mainline subclasses. The basic form of the number of mainline subclasses of the 13 classes is 159,
a citation frequency matrix is shown in Table 2 where fNiMj the resulting citation frequency matrix is a 159  159 matrix,
denotes the number of patents in ith mainline subclass of but not shown here due to the space limit. Only the citation
class N that patents in jth mainline subclass of class M cite. frequency matrix at the class level is shown in Table 3.

4.4. ANP
Table 1
Telecommunication technology classes 4.4.1. Network model construction
Class Title Number of mainline Basically, a network model in ANP is constructed based
subclasses on expert judgments to model an abstract decision prob-
329 Demodulators 7 lem. However, the network in the proposed approach is
331 Oscillators 37 made on the basis of citation relationships represented in
332 Modulators 7
the citation frequency matrix, as is in the case of network
340 Communications: electrical 13
341 Coded data generation or conversion 10 analysis. A cluster in the ANP network corresponds to a
342 Communications: directive radio wave 11 class, and elements in a cluster are equivalent to mainline
systems and devices subclasses in a class. In the ANP context, then, the result-
343 Communications: radio wave antennas 1 ing network model only includes alternative clusters, con-
367 Communications, electrical: acoustic wave 3
trary to the general network model in the ANP
systems and devices
370 Multiplex communications 12 comprised of a goal cluster, criteria clusters, and alternative
375 Pulse or digital communications 20 clusters. Thus, the importance of alternatives is only evalu-
379 Telephonic communications 16 ated with respect to impacts or influences on other alterna-
380 Cryptography 15 tives, not with respect to criteria or a goal, which is the
455 Telecommunications 7
same as the idea of centrality measures in network analysis.
900 H. Lee et al. / Expert Systems with Applications 36 (2009) 894–908

Table 3
Citation frequency matrix at class level
455 380 379 375 370 367 343 342 341 340 332 331 329
455 15,487 7714 166 6804 744 20 10 344 140 1034 2837 12,884 3583
380 3303 78,954 397 1160 794 3 0 32 241 1369 114 273 176
379 1521 4514 400 208 212 36 0 26 17 279 77 200 37
375 6469 5030 29 19,587 1380 50 4 226 356 294 4466 9605 7497
370 1815 6560 58 3151 1827 35 0 54 61 310 540 1058 1229
367 59 87 0 11 0 4 0 8 55 17 3 104 15
343 94 46 0 46 0 0 128 43 0 17 16 386 18
342 844 693 3 580 33 5 18 860 4 295 191 1347 405
341 284 1132 26 770 20 12 0 12 782 227 208 1882 237
340 2145 12,084 81 644 141 0 24 300 482 5211 130 1379 314
332 1727 51 2 2630 53 5 0 33 74 15 2620 2869 509
331 10,748 566 9 7180 104 70 57 495 207 137 2045 101,992 1121
329 1495 34 0 3384 80 5 0 63 47 40 453 1373 2435

An arrow indicates the existence of patent citation rela-
tionships between classes or mainline subclasses. For exam-
ple, an arrow which leaves class A and enters into class B is
added to a network if some of the patents in class A cite
some of the patents in class B. What this also means is class
B has some influences on class A; thus, subclasses of class B
should be pair-wisely compared with respect to impacts on
each subclass of class A.
Fig. 4 shows the telecommunication technology network
for ANP including the 13 classes. Every class has influences
on each other, and includes a feedback loop that represents
citation relationships among mainline subclasses in the
class itself. Though the network can be elaborated more
by describing citation relationships at the mainline class
level, it is not represented due to its complexity.
4.4.2. Pairwise comparisons and priority vectors
The next step deals with obtaining priority vectors.
Firstly, cluster weights are determined through comparisons
at the cluster level. The basic form of measurement in the
ANP is a pairwise comparison with a scale of 1–9 since sub-
ject judgments have to be made on qualitative aspects.
However, pairwise comparisons do not have to be done
in the proposed approach. It is implicitly assumed that the
number of patent citations between a pair of nodes is a
proxy of intensity of influence. Then, the importance of
elements can be directly measured from the citation fre-
quency matrix. For example, Table 3 shows that the num-
ber of citations made by patents of class 455 is 3303 for
the patents of class 380, 1521 for the patents of class
379. This can be interpreted that class 380 is about

Fig. 4. Telecommunication technology network for ANP.

 H. Lee et al. / Expert Systems with Applications 36 (2009) 894–908 901

2.17 (=3303/1.521) times more important than class 379 in is assigned to position (375, 370). In this way, the pairwise
terms of impacts on class 455. Then, the number 2.17 is comparison matrix with respect to class 455 among the 13
inserted to position (380, 379) and reciprocal value, 0.46, classes can be obtained as shown in Table 4. Then, the

Table 4
Pairwise comparison matrix with respect to class 455 and resulting priority vector
h455i 455 380 379 ... 331 329 Priority Normalization
455 1 4.69 10.18 1.44 10.36 0.3367 =15,487/45,991
380 1 2.17 0.31 2.21 0.0718 =3303/45,991
379 1 0.14 1.02 0.0331 =1521/45,991
375 0.60 4.33 0.1407 =6469/45,991
370 0.17 1.21 0.0395 =1815/45,991
367 0.01 0.04 0.0013 =59/45,991
343 0.01 0.06 0.0020 =94/45,991
342 0.08 0.56 0.0184 =844/45,991
341 0.03 0.19 0.0062 =284/45,991
340 0.20 1.43 0.0466 =2145/45,991
332 0.16 1.16 0.0376 =1727/45,991
331 1 7.19 0.2337 =10,748/45,991
329 1 0.0325 =1495/45,991

Table 5
Cluster weights
455 380 379 375 370 367 343 342 341 340 332 331 329
455 0.3367 0.0657 0.1418 0.1474 0.1381 0.0816 0.0415 0.1378 0.0568 0.1118 0.2071 0.0952 0.2039
380 0.0718 0.6721 0.3390 0.0251 0.1474 0.0122 0.0000 0.0128 0.0977 0.1481 0.0083 0.0020 0.0100
379 0.0331 0.0384 0.3416 0.0045 0.0393 0.1469 0.0000 0.0104 0.0069 0.0302 0.0056 0.0015 0.0021
375 0.1407 0.0428 0.0248 0.4244 0.2561 0.2041 0.0166 0.0905 0.1444 0.0318 0.3260 0.0710 0.4265
370 0.0395 0.0558 0.0495 0.0683 0.3391 0.1429 0.0000 0.0216 0.0247 0.0335 0.0394 0.0078 0.0699
367 0.0013 0.0007 0.0000 0.0002 0.0000 0.0163 0.0000 0.0032 0.0223 0.0018 0.0002 0.0008 0.0009
343 0.0020 0.0004 0.0000 0.0010 0.0000 0.0000 0.5311 0.0172 0.0000 0.0018 0.0012 0.0029 0.0010
342 0.0184 0.0059 0.0026 0.0126 0.0061 0.0204 0.0747 0.3446 0.0016 0.0319 0.0139 0.0100 0.0230
341 0.0062 0.0096 0.0222 0.0167 0.0037 0.0490 0.0000 0.0048 0.3171 0.0246 0.0152 0.0139 0.0135
340 0.0466 0.1029 0.0692 0.0140 0.0262 0.0000 0.0996 0.1202 0.1955 0.5637 0.0095 0.0102 0.0179
332 0.0376 0.0004 0.0017 0.0570 0.0098 0.0204 0.0000 0.0132 0.0300 0.0016 0.1912 0.0212 0.0290
331 0.2337 0.0048 0.0077 0.1556 0.0193 0.2857 0.2365 0.1983 0.0839 0.0148 0.1493 0.7535 0.0638
329 0.0325 0.0003 0.0000 0.0733 0.0148 0.0204 0.0000 0.0252 0.0191 0.0043 0.0331 0.0101 0.1385

Table 6
Citation frequency matrix and its transformation into priority matrix
455
3.01 403 7 39 73 91 130
Citation frequency matrix
329 300 0 0 2 3 30 13 152
304 0 1 3 16 38 37 347
311 0 0 0 4 3 6 27
315 0 7 0 17 64 27 471
345 0 1 0 3 2 5 30
347 0 0 0 12 24 20 129
372 0 0 0 0 0 0 1
Priority matrix
329 300 0.0000 0.0000 0.4000 0.0545 0.1863 0.1204 0.1314
304 0.0000 0.1111 0.6000 0.2909 0.2360 0.3426 0.2999
311 0.0000 0.0000 0.0000 0.0727 0.0186 0.0556 0.0233
315 0.0000 0.7778 0.0000 0.3091 0.3975 0.2500 0.4071
345 0.0000 0.1111 0.0000 0.0545 0.0124 0.0463 0.0259
347 0.0000 0.0000 0.0000 0.2182 0.1491 0.1852 0.1115
372 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0009
902 H. Lee et al. / Expert Systems with Applications 36 (2009) 894–908

priority vector for class 455 is derived from the eigenvec-

0.0000
0.0938
0.0000
0.0625
0.1875
0.0625
0.5938

0.1250
0.2500
0.0000
0.3750
0.1250
0.1250
0.0000
372
tor method. This priority vector is naturally the same as

.
.
.
.
.
.
.
.
.
.
.
the vector of the number of citations that each of 13

0.0000
0.0083
0.0062
0.0922
0.0598
0.1774
0.6561

0.1010
0.1838
0.0505
0.2929
0.0222
0.3394
0.0014
347
classes received divided by total number of citations made

.
.
.
.
.
.
.
.
.
.
.
by the patents of class 455. This is because the pairwise

0.0000
0.0208
0.0000
0.0625
0.0417
0.0208
0.8542

0.0735
0.6029
0.0441
0.1618
0.0588
0.0441
0.0020
345
comparison matrix is a completely consistent matrix.

.
.
.
.
.
.
.
.
.
.
.
Therefore, the priority vectors can be directly obtained

0.0000
0.0096
0.0032
0.0777
0.0714
0.0469
0.7913

0.0854
0.1357
0.0244
0.6113
0.0442
0.0976
0.0002
315
from the citation frequency matrix without pairwise

.
.
.
.
.
.
.
.
.
.
.
comparisons.

0.0000
0.0517
0.0000
0.1379
0.0000
0.1034
0.7069

0.0000
0.2500
0.5167
0.1000
0.0000
0.1333
0.0000
Table 5 shows the priority vectors for each cluster,

311

.
.
.
.
.
.
.
.
.
.
.
that is, the cluster weights derived in this way. The clus-

0.0011
0.0504
0.0258
0.1153
0.0582
0.0582
0.6909

0.0498
0.7658
0.0102
0.0905
0.0158
0.0645
0.0005
ter weights will be used to obtain the weighted super-

304

.
.
.
.
.
.
.
.
.
.
.
matrix.

0.0000
0.0126
0.0126
0.0755
0.0943
0.0377
0.7673

0.5038
0.1515
0.0114
0.2879
0.0303
0.0114
0.0005
Secondly, local priority vectors for mainline subclasses

329

300

.
.
.
.
.
.
.
.
.
.
.
are obtained. In ANP, basically, pairwise comparisons
are made among elements of a cluster an arrow enters with

331
respect to each element of a cluster from which an arrow

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
leaves. For a feedback loop, elements in a cluster are

332
pair-wisely compared with respect to each element in the

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
cluster itself. For each pairwise comparison supposed to

340
be made, local priority vectors can be directly derived

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
without pairwise comparisons as mentioned above. For 341
example, the importance of mainline subclasses of class .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
370 on each mainline subclass of class 450 is obtained by
transformation of the citation frequency matrix, as shown
342

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
in Table 6. What is important here is normalization of col-
umns has to be done for each cluster. The resulting set of
343

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
priority vectors, a priority matrix, will be imported to the
supermatrix.
367

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
370

4.4.3. Supermatrix formation and transformation

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
The supermatrix is constructed with local priority vec-
375

tors obtained from the previous step. The supermatrix for

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
the telecommunication technology network, which is a
159  159 matrix composed of 169 (=13  13) blocks. A
379

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
block corresponds to a set of priority vectors, a priority
matrix. The priority matrix in Table 6 is equivalent to
380

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
W13,1 in the supermatrix. Table 7 shows a part of the limit
0.0011
0.0089
0.0086
0.0479
0.1020
0.1006
0.7310

0.1314
0.2999
0.0233
0.4071
0.0259
0.1115
supermatrix. 0.0009
130

.
.
.
.
.
.
.
.
.
.
.

The supermatrix then needs to be transformed into the

0.0023
0.0320
0.0107
0.0846
0.1524
0.5290
0.1890

0.1204
0.3426
0.0556
0.2500
0.0463
0.1852
0.0000

weighted supermatrix. Each matrix segment of the

91

.
.
.
.
.
.
.
.
.
.
.

supermatrix is multiplied by the corresponding cluster

0.0005
0.1625
0.0286
0.0561
0.2166
0.1162
0.4195

0.1863
0.2360
0.0186
0.3975
0.0124
0.1491
0.0000

weights shown in Table 5. For example, all the elements

73

of W11 are multiplied by the weight of class 455 for class

.
.
.
.
.
.
.
.
.
.
.
0.0051
0.0925
0.0231
0.2828
0.0874
0.0951
0.4139

0.0545
0.2909
0.0727
0.3091
0.0545
0.2182
0.0000

455 itself, 0.3367, W13,1 is multiplied by 0.0325, and so on.

39

However, the resulting matrix is not column stochastic

.
.
.
.
.
.
.
.
.
.
.
0.0000
0.3371
0.1873
0.1161
0.2472
0.0749
0.0375

0.4000
0.6000
0.0000
0.0000
0.0000
0.0000
0.0000

because there are several matrix segments that have col-

umns all of whose entries are zero. When this is the case,
7

.
.
.
.
.
.
.
.
.
.
.

the weighted column of the supermatrix must be renor-

0.0005
0.6938
0.0371
0.0739
0.1554
0.0075
0.0318

0.0000
0.1111
0.0000
0.7778
0.1111
0.0000
0.0000
403

malized (Saaty, 1996). The renormalized matrix, which is

.
.
.
.
.
.
.
.
.
.
.

now column stochastic, is what is called the weighted

0.0000
0.1379
0.0000
0.2069
0.3793
0.1379
0.1379

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
3.01
455

supermatrix. A part of the weighted supermatrix is shown

.
.
.
.
.
.
.
.
.
.
.

in Table 8. Finally, the limit supermatrix was derived by

3.01
403

130

300
304
311
315
345
347
372
Supermatrix

raising the weighted supermatrix to powers. In this case,

39
73
91
7

.
.
.
.
.
.
.
.
.
.
.

convergence is reached at W41. Table 9 shows a part of

Table 7

455

380
379
375
370
367
343
342
341
340
332
331
329

the limit supermatrix.

Table 8
Weighted supermatrix

455 380 379 375 370 367 343 342 341 340 332 331 329

3.01 403 7 39 73 91 130 . . . . . . . . . . . 300 304 311 315 345 347 372
455 3.01 0.0000 0.0002 0.0000 0.0017 0.0002 0.0008 0.0004 . . . . . . . . . . . 0.0000 0.0002 0.0000 0.0000 0.0000 0.0000 0.0000
403 0.0695 0.2339 0.1144 0.0312 0.0547 0.0108 0.0030 . . . . . . . . . . . 0.0026 0.0103 0.0105 0.0020 0.0043 0.0017 0.0221
7 0.0000 0.0125 0.0635 0.0078 0.0096 0.0036 0.0029 . . . . . . . . . . . 0.0026 0.0053 0.0000 0.0007 0.0000 0.0013 0.0000
39 0.1042 0.0249 0.0394 0.0953 0.0189 0.0285 0.0161 . . . . . . . . . . . 0.0154 0.0235 0.0281 0.0159 0.0128 0.0188 0.0147
73 0.1911 0.0524 0.0839 0.0295 0.0730 0.0513 0.0343 . . . . . . . . . . . 0.0193 0.0119 0.0000 0.0146 0.0085 0.0122 0.0442
91 0.0695 0.0025 0.0254 0.0321 0.0391 0.1781 0.0339 . . . . . . . . . . . 0.0077 0.0119 0.0211 0.0096 0.0043 0.0362 0.0147
130 0.0695 0.0107 0.0127 0.1395 0.1413 0.0637 0.2461 . . . . . . . . . . . 0.1570 0.1409 0.1441 0.1615 0.1748 0.1338 0.1400
380 . . . . . . . . . . . . . . . . . . . . . . . . . .
379 . . . . . . . . . . . . . . . . . . . . . . . . . .
375 . . . . . . . . . . . . . . . . . . . . . . . . . .
370 . . . . . . . . . . . . . . . . . . . . . . . . . .
367 . . . . . . . . . . . . . . . . . . . . . . . . . .
343 . . . . . . . . . . . . . . . . . . . . . . . . . .
342 . . . . . . . . . . . . . . . . . . . . . . . . . .

H. Lee et al. / Expert Systems with Applications 36 (2009) 894–908

341 . . . . . . . . . . . . . . . . . . . . . . . . . .
340 . . . . . . . . . . . . . . . . . . . . . . . . . .
332 . . . . . . . . . . . . . . . . . . . . . . . . . .
331 . . . . . . . . . . . . . . . . . . . . . . . . . .
329 300 0.0000 0.0000 0.0131 0.0018 0.0061 0.0039 0.0043 . . . . . . . . . . . 0.0701 0.0069 0.0000 0.0118 0.0102 0.0140 0.0200
304 0.0000 0.0036 0.0197 0.0095 0.0077 0.0111 0.0097 . . . . . . . . . . . 0.0211 0.1061 0.0346 0.0188 0.0839 0.0255 0.0401
311 0.0000 0.0000 0.0000 0.0024 0.0006 0.0018 0.0008 . . . . . . . . . . . 0.0016 0.0014 0.0716 0.0034 0.0061 0.0070 0.0000
315 0.0000 0.0253 0.0000 0.0101 0.0129 0.0081 0.0132 . . . . . . . . . . . 0.0400 0.0125 0.0139 0.0848 0.0225 0.0406 0.0601
345 0.0000 0.0036 0.0000 0.0018 0.0004 0.0015 0.0008 . . . . . . . . . . . 0.0042 0.0022 0.0000 0.0061 0.0082 0.0031 0.0200
347 0.0000 0.0000 0.0000 0.0071 0.0048 0.0060 0.0036 . . . . . . . . . . . 0.0016 0.0089 0.0185 0.0135 0.0061 0.0470 0.0200
372 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 . . . . . . . . . . . 0.0005 0.0005 0.0000 0.0002 0.0020 0.0014 0.0000

Table 9
Limit supermatrix

455 380 379 375 370 367 343 342 341 340 332 331 329

3.01 403 7 39 73 91 130 . . . . . . . . . . . 300 304 311 315 345 347 372
455 3.01 0.0003 0.0003 0.0003 0.0003 0.0003 0.0003 0.0003 . . . . . . . . . . . 0.0003 0.0003 0.0003 0.0003 0.0003 0.0003 0.0003
403 0.0194 0.0193 0.0194 0.0194 0.0194 0.0194 0.0194 . . . . . . . . . . . 0.0193 0.0193 0.0193 0.0193 0.0193 0.0193 0.0193
7 0.0032 0.0032 0.0032 0.0032 0.0032 0.0032 0.0032 . . . . . . . . . . . 0.0032 0.0032 0.0032 0.0032 0.0032 0.0032 0.0032
39 0.0164 0.0164 0.0164 0.0164 0.0164 0.0164 0.0164 . . . . . . . . . . . 0.0164 0.0164 0.0164 0.0164 0.0164 0.0164 0.0164
73 0.0174 0.0173 0.0174 0.0174 0.0174 0.0174 0.0174 . . . . . . . . . . . 0.0174 0.0173 0.0174 0.0174 0.0174 0.0174 0.0174
91 0.0180 0.0180 0.0180 0.0180 0.0180 0.0180 0.0180 . . . . . . . . . . . 0.0180 0.0180 0.0180 0.0180 0.0180 0.0180 0.0180
130 0.0699 0.0699 0.0699 0.0699 0.0699 0.0699 0.0699 . . . . . . . . . . . 0.0699 0.0699 0.0699 0.0699 0.0699 0.0699 0.0699
380 . . . . . . . . . . . . . . . . . . . . . . . . . .
379 . . . . . . . . . . . . . . . . . . . . . . . . . .
375 . . . . . . . . . . . . . . . . . . . . . . . . . .
370 . . . . . . . . . . . . . . . . . . . . . . . . . .
367 . . . . . . . . . . . . . . . . . . . . . . . . . .
343 . . . . . . . . . . . . . . . . . . . . . . . . . .
342 . . . . . . . . . . . . . . . . . . . . . . . . . .
341 . . . . . . . . . . . . . . . . . . . . . . . . . .
340 . . . . . . . . . . . . . . . . . . . . . . . . . .
332 . . . . . . . . . . . . . . . . . . . . . . . . . .
331 . . . . . . . . . . . . . . . . . . . . . . . . . .
329 300 0.0025 0.0025 0.0025 0.0025 0.0025 0.0025 0.0025 . . . . . . . . . . . 0.0025 0.0025 0.0025 0.0025 0.0025 0.0025 0.0025
304 0.0102 0.0102 0.0102 0.0102 0.0102 0.0102 0.0102 . . . . . . . . . . . 0.0102 0.0102 0.0102 0.0102 0.0102 0.0102 0.0102
311 0.0015 0.0015 0.0015 0.0015 0.0015 0.0015 0.0015 . . . . . . . . . . . 0.0015 0.0015 0.0015 0.0015 0.0015 0.0015 0.0015
315 0.0075 0.0075 0.0075 0.0075 0.0075 0.0075 0.0075 . . . . . . . . . . . 0.0075 0.0075 0.0075 0.0075 0.0075 0.0075 0.0075
345 0.0008 0.0008 0.0008 0.0008 0.0008 0.0008 0.0008 . . . . . . . . . . . 0.0008 0.0008 0.0008 0.0008 0.0008 0.0008 0.0008
347 0.0026 0.0026 0.0026 0.0026 0.0026 0.0026 0.0026 . . . . . . . . . . . 0.0026 0.0026 0.0026 0.0026 0.0026 0.0026 0.0026
372 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 . . . . . . . . . . . 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000

903
904 H. Lee et al. / Expert Systems with Applications 36 (2009) 894–908

Table 10 from the USPTO, the ANP network model was con-
Limit centralities of 13 classes structed and local priority vectors were obtained. Forming
No. Title Limit and transforming the supermatrix led to converged priori-
centrality ties, limit centralities.
455 Telecommunications 0.0251 The main contribution of this study is to apply the
380 Cryptography 0.2885 MCDM methodology, ANP, to a technology network.
379 Telephonic communications 0.0272
375 Pulse or digital communications 0.0883
Since ANP captures the relative importance that mirrors
370 Multiplex communications 0.0191 all the direct and indirect interactions, the limit centrality
367 Communications, electrical: acoustic wave systems 0.0219 measures importance of technologies in terms of impacts
and devices on other technologies in the technology network, taking
343 Communications: radio wave antennas 0.0034 indirect impacts or relationships into account, which is very
342 Communications: directive radio wave systems and 0.0023
devices
difficult or tedious with the conventional centrality mea-
341 Coded data generation or conversion 0.0515 sures. The applicability of limit centrality is not limited
340 Communications: electrical 0.1547 to a technology network. For any type of social networks,
332 Modulators 0.0263 the limit centrality can be used as an implicative centrality
331 Oscillators 0.1464 measure characterizing a network and showing core actors
329 Demodulators 0.1446
in the network.
Nevertheless, this research is still subject to some limi-
tations. Firstly, it cannot be used for undirectional net-
4.5. Limit centrality works where an edge has no direction and only
represents the existence of a relationship between two
As the supermatrix covers the whole network, the col- nodes since relationships in a network of ANP must have
umns in the limit supermatrix (Table 9) represent final pri- directions depending on the influence between elements or
orities, namely, limit centralities. That is why it is called clusters. Secondly, the influences among patent classes are
limit centrality. Due to the nature of limit priorities in measured by the absolute size of patent citations; thus, we
ANP, the limit centralities of all the elements sum to one. cannot control the effect of the size of a class, that is, total
The limit centrality indicates importance of technologies number of patents in a class, on measuring the degree of
in terms of impacts on other technologies, taking all the impacts. The relative impact may be a more implicative
direct and indirect influences into consideration. The limit measure depending on the context. It can be derived by
centralities of the 159 mainline subclasses are shown in dividing each column of the citation frequency matrix
Appendix B. The limit centrality of a class is the sum of by the total number of patents in the corresponding class.
mainline subclasses belonging to the class. Table 10 shows Thirdly, the selected 13 patent classes as telecommunica-
the limit centralities of 13 classes. tion technologies are by no means exhaustive. A more
At the mainline subclass level, the one with the highest systematic procedure is required to select the target
limit centrality is 455/130 (Receiver or analog modulated classes.
signal frequency converter), and the next is 375/316 These limitations could serve as fruitful avenues for
(Receivers). It is obvious that these technologies have sig- future research. Applications of the proposed approach
nificant impacts on other technologies, and therefore they to a variety of networks can be a worthwhile area for future
are considered as the core technologies of the telecommuni- research. A dynamic analysis on the telecommunication
cation technology network. When it comes to the class network is also expected to provide useful information on
level, 380 (Cryptography) has the highest limit centrality, the change of the network structure and technological
followed by 340 (Communications: electrical), 331 (Oscilla- trends.
tors), and 329 (Demodulators).
On the other hand, the limit centrality of 379/414 (trans-
mission line conditioning) is zero since its patents have
never been cited by all the patents of the other classes. Appendix A. ICT classification and corresponding IPC codes
The class whose limit centrality is the lowest is 342 (Com-
munications: directive radio wave systems and devices).
Classifications IPC codes
5. Conclusions Telecommunication G01S, G08C, G09C, H01P, H01Q,
H01S3/(025, 243, 063, 067, 085,
The proposed approach measures the limit centralities 0933, 0941, 103, 133, 18, 19, 25),
of technologies with the aim of identification of core tech- H1S5, H03B, H03C, H03D, H03H,
nologies in the technology network. A case study on the H03M, H04B, H04J, H04K, H04L,
telecommunication technology network was presented to H04M, H04Q
illustrate the proposed approach. After constructing the Consumer G11B, H03F, H03G, H03J, H04H,
citation frequency matrix based on patent data collected electronics H04N, H04R, H04S
 H. Lee et al. / Expert Systems with Applications 36 (2009) 894–908 905

Appendix A (continued) Appendix B (continued)

Classifications IPC codes Class Title Limit
Computers, B07C, B41J, B41K, G02F, G03G, G05F, no. centrality
office G06, G07, G09G, G10L, G11C, H03K, 56 Selectively movable element having 0.0002
machinery H03L code characters
Other ICT G01B, G01C, G01D, G01F, G01G, 59 Miscellaneous 0.0008
G01H, G01J, G01K, G01L, G01M,
G01N, G01P, G01R, G01V, G01W, 379 Telephonic communications 0.0272
G02B6, G05B, G08G, G09B, H01B11, 1.01 Diagnostic testing, malfunction 0.0005
H01J(11/, 13/, 15/, 17/, 19/, 21/, 23/, 25/, indication, or electrical condition
27/, 29/, 31/, 33/, 40/, 41/, 43/, 45/), H01L measurement
67.1 Audio message storage, retrieval, or 0.0031
synthesis
90.01 Telephone line or system combined 0.0106
with diverse electrical system or
Appendix B. One hundred and fifty nine mainline subclasses’
signalling (e.g., composite)
titles and derived limit centralities
110.01 Composite substation or therminal 0.0001
(e.g., having calculator, radio)
111 With usage measurement (e.g., call or 0.0022
Class Title Limit traffic register)
no. centrality 142.01 Reception of calling information at 0.0004
455 Telecommunications 0.0251 substation in wireline
3.01 Wireless distribution system 0.0003 communications system
403 Radiotelephone system 0.0194 156 Multi-line or key substation system 0.0003
7 Carrier wave repeater or relay system 0.0032 with selective switching and central
(i.e., retransmission of same switching office connection
information) 188 Call or terminal access alarm or 0.0009
39 Transmitter and receiver at separate 0.0164 control
stations 201.01 Special services 0.0017
73 Transmitter and receiver at same 0.0174 219 Plural exchange network or 0.0005
station (e.g., transceiver) interconnection
91 Transmitter 0.0180 242 Centralized switching system 0.0011
130 Receiver or analog modulated signal 0.0699 350 Supervisory or control line signaling 0.0018
frequency converter 399.01 Subscriber line or transmission line 0.0010
interface
380 Cryptography 0.2885 414 Transmission line conditioning 0.0000
1 Cryptanalysis 0.0002 419 Terminal 0.0016
2 Equipment test or malfunction 0.0009 441 Terminal accessory or auxiliary 0.0004
indication 1 equipment
200 Video cryptography 0.0216
243 Facsimile cryptography 0.0018 375 Pulse or digital communications 0.0883
247 Cellular telephone cryptographic 0.0050 130 Spread spectrum 0.0162
authentication 211 Repeaters 0.0002
251 Electronic game using cryptography 0.0001 216 Apparatus convertible to analog 0.0013
252 Electric signal masking 0.0006 219 Transceivers 0.0042
255 Communication system using 0.0422 224 Testing 0.0013
cryptography 229 Equalizers 0.0042
277 Key management 0.0346 237 Pulse number modulation 0.0001
28 Particular algorithmic function 0.0292 238 Pulse with modulation 0.0011
encoding 239 Pulse position, frequency, or spacing 0.0011
287 Electric signal modification 0.0059 modulation
54 By modifying optical image (e.g., 0.0023 240 Bandwidth reduction or expansion 0.0039
transmissive overlay) 242 Pulse code modulation 0.0016
55 Having production of printed copy 0.0012 256 Pulse transmission via radiated 0.0003
(e.g., cryptographic printer or baseband
typerwriter) (continued on next page)
906 H. Lee et al. / Expert Systems with Applications 36 (2009) 894–908

Appendix B (continued) Appendix B (continued)

Class Title Limit Class Title Limit
no. centrality no. centrality
257 Cable systems and components 0.0004 341 Coded data generation or conversion 0.0515
259 Systems using alternating or pulsating 0.0165 1 Digital pattern reading type converter 0.0008
current 20 Bodily actuated code generator 0.0016
286 Multilevel 0.0014 50 Digital code to digital code converters 0.0034
295 Transmitters 0.0160 110 Analog to digital conversion followed 0.0003
316 Receivers 0.0497 by digital to analog conversion
353 Pulse amplitude modulation 0.0002 111 Phase or time of phase change 0.0002
354 Synchronizers 0.0344 118 Converter compensation 0.0009
377 Miscellaneous 0.0007 120 Converter calibration or testing 0.0010
122 Sample and hold 0.0006
370 Multiplex communications 0.0191 126 Analog to or from digital conversion 0.0068
203 Generalized orthogonal or special 0.0051 173 Code generator or transmitter 0.0034
mathematical techniques
212 Pulse width (pulse duration) 0.0002 340 Communications: electrical 0.1547
modulation 870.01 Continuously variable indicating (e.g., 0.0043
213 Pulse position modulation 0.0004 telemetering)
215 Phase modulation 0.0001 901 External condition vehicle-mounted 0.0008
229 Data flow congestion prevention or 0.0008 indicator or alarm
control 907 Traffic control indicator 0.0006
241 Diagnostic testing (other than 0.0009 933 Vehicle detectors 0.0010
synchronization) 988 Vehicle position indication 0.0023
259 Special services 0.0004 425.5 Land vehicle alarms or indicators 0.0056
276 Duplex 0.0024 500 Condition responsive indicating 0.0122
310 Communication over free space 0.0133 system
351 Pathfinding or routing 0.0056 825 Selective 0.0586
431 Channel assignment techniques 0.0031 310.11 Remote control over power line 0.0006
464 Communication techniques for 0.0191 286.01 Systems 0.0006
information carried in plural channels 407.1 Tactual indication 0.0001
815.4 Visual indication 0.0001
367 Communications, electrical: acoustic 0.0219 384.1 Audible indication 0.0012
wave systems and devices
14 Seismic prospecting 0.0003 332 Modulators 0.0263
87 Echo systems 0.0004 100 Frequency shift keying modulator or 0.0031
140 Signal transducers 0.0023 minimum shift keying modulator
103 Phase shift keying modulator or 0.0055
343 Communications: radio wave antennas 0.0034 quadrature amplitude modulator
700r Antennas 0.0034 106 Pulse or interrupted continuous wave 0.0022
modulator
342 Communications: directive radio wave 0.0023 117 Frequency modulator 0.0108
systems and devices 144 Phase modulator 0.0019
13 Radar ew (electronic warfare) 0.0005 149 Amplitude modulator 0.0037
21 Base band system 0.0003 185 Miscellaneous 0.0000
27 Radar for meteorological use (EPO) 0.0003
42 Radar transponder system 0.0065 331 Oscillators 0.1464
61 Return signal controls external device 0.0003 94.1 Molecular or particle resonant type 0.0006
73 Return signal controls radar system 0.0009 (e.g., maser)
118 Determining distance 0.0014 1r Automatic frequency stabilization 0.1589
159 Clutter elimination 0.0002 using a phase or frequency sensing
165 Testing or calibrating of radar system 0.0007 means
175 With particular circuit 0.0024 37 Beat frequency 0.0020
350 Directive 0.0083 44 With frequency calibration or testing 0.0019
 H. Lee et al. / Expert Systems with Applications 36 (2009) 894–908 907

Appendix B (continued) Appendix B (continued)

Class Title Limit Class Title Limit
no. centrality no. centrality
45 Polyphase output 0.0018 315 Frequency modulation 0.0075
46 Plural oscillators 0.0079 demodulator
57 Ring oscillators 0.0107 345 Phase modulation demodulator 0.0008
59 Convertible (e.g., oscillator to 0.0001 347 Amplitude modulation 0.0026
amplifier, etc.) demodulator
60 Single oscillator with plural output 0.0010 372 Miscellaneous 0.0000
circuits
62 With oscillator circuit protective 0.0001
means
64 With indicator, signal, or alarm 0.0003 References
65 With device responsive to external 0.0025
physical condition Agarwal, A., Shankar, R., & Tiwari, M. K. (2006). Modeling the metrics
67 With electromagnetic or electrostatic 0.0001 of lean, agile and leagile supply chain: An ANP-based approach.
shield European Journal of Operational Research, 173(1), 211–225.
68 With outer casing or housing 0.0012 Archibugi, D., & Pianta, M. (1996). Measuring technological change
through patents and innovation surveys. Technovation, 16, 451–458.
70 With temperature modifier 0.0002 Ayag, Z., & Özdemir, R. G. (2007). An analytic network process-based
72 Electron-coupled type 0.0000 approach to concept evaluation in a new product development
74 Combined with particular output 0.0059 environment. Journal of Engineering Design, 18(3), 209–226.
coupling network Basberg, B. L. (1984). Patent statistics and the measurement of techno-
78 Electrical noise or random wave 0.0034 logical change: An assessment of the Norwegian patent data, 1840–
1980. World Patent Information, 6, 158–164.
generator Basberg, B. L. (1987). Patents and the measurement of technological
79 Beam tube 0.0000 change: A survey of literature. Research Policy, 16, 131–141.
86 With magnetically controlled space 0.0001 Bayazit, O., & Karpak, B. (2007). An analytical network process-based
discharge device (e.g., magnetron) framework for successful total quality management (TQM): An
96 With distributed parameter resonator 0.0050 assessment of Turkish manufacturing industry readiness. International
Journal of Production Economics, 105(1), 79–96.
105 With parasitic oscillation control or 0.0003 Bonacich, P., & Lloyd, P. (2001). Eigenvector-like measures of centrality
prevention means for asymmetric relations. Social Networks, 23(3), 191–201.
107r Solid state active element oscillator 0.0319 Bonacich, P. (1972). Factoring and weighting approaches to status scores
126 Gaseous space discharge device 0.0000 and clique identification. Journal of Mathematical Sociology, 2,
132 Negative resistance or negative 0.0004 113–120.
Borgatti, S. P. (2005). Centrality and network flow. Social Networks,
transconductance oscillator 27(1), 55–71.
135 Phase shift type 0.0011 Breitzman, A., & Thomas, P. (2002). Using patent citation analysis to
138 Bridge type 0.0002 target/value M&A candidates. Research Technology Management, 45,
143 Relaxation oscillators 0.0027 28–46.
154 Electromechanical resonator 0.0060 Breschi, S., Lissoni, F., & Maleraba, F. (1998). Knowledge proximity and
technological diversification, CESPRI, ISE Research Project.
165 Shock excited resonant circuit 0.0002 Chang, C. W., Wu, C. R., Lin, C. T., & Lin, H. L. (2007). Evaluating
167 L-C type oscillators 0.0015 digital video recorder systems using analytic hierarchy and analytic
172 With synchronizing, triggering or 0.0019 network processes. Information Sciences, 177(16), 3383–3396.
pulsing circuits Chen, D.-Z., Chang, H.-W., Huang, M.-H., & Fu, F.-C. (2005). Core
175 Frequency stabilization 0.0078 technologies and key industries in Taiwan from 1978 to 2002: A
perspective from patent analysis. Scientometrics, 64, 31–53.
177r With frequency adjusting means 0.0263 Cheng, E. W. L., & Li, H. (2005). Analytic network process applied to
182 Amplitude control or stabilization 0.0018 project selection. Journal of construction engineering and management,
185 With particular source of power or 0.0026 131(4), 459–466.
bias voltage Chung, S., Lee, A., & Pearn, W. (2005). Analytic network process (ANP)
187 Miscellaneous oscillator structures 0.0001 approach for product mix planning in semiconductor fabricator.
International Journal of Production Economics, 96(1), 15–36.
Courtial, J. P., Callon, M., & Sigogneau, A. (1993). The use of patent titles
329 Demodulators 0.1446 for identifying the topics of invention and forecasting trends. Sciento-
300 Frequency shift keying or minimum 0.0025 metrics, 26(2), 231–242.
shift keying demodulator Demirtas, E. A., & Ustun, O. (2007). Analytic network process and multi-
304 Phase shift keying or quadrature 0.0102 period goal programming integration in purchasing decisions. Com-
puters & Industrial Engineering, doi:10.1016/j.cie.2006.12.006.
amplitude demodulator EIRMA (2000). Technology monitoring for business success. European
311 Pulse or interrupted continuous wave 0.0015 Industrial Research Management Association, Working Group 55
demodulator Report.
908 H. Lee et al. / Expert Systems with Applications 36 (2009) 894–908
Ernst, H. (2003). Patent information for strategic technology manage-
ment. World Patent Information, 25(3), 233–242.
Freeman, L. C. (1979). Centrality in social networks: Conceptual
clarification. Social Networks, 1, 215–239.
Gangulli, P. (2004). Patents and patent information in 1979 and 2004: A
perspective from India. World Patent Information, 26, 61–62.
Gelsing, L. (1992). Innovation and the development of industrial
networks. In B.-A. Lundvall (Ed.), National systems of innovation –
Towards a theory of innovation and interactive learning (pp. 116–128).
London: Pinter.
Gencer, C., & Gürpinar, D. (2007). Analytic network process in supplier
selection: A case study in an electronic firm. Applied Mathematical
Modelling, 31(11), 2475–2486.
Grilliches, Z. (1990). Patent statistics as economic indicators: A survey.
Journal of Economic Literature, 28, 1661–1707.
Grupp, H. (1996). Spillover effects and the science base of innovations
reconsidered: An empirical approach. Journal of Evolutionary Eco-
nomics, 6(2), 175–197.
Grupp, H., Lacasa, D., & Schmoch, U. (2003). Tracing technological
change over long periods in Germany in chemicals using patent
statistics. Scientometrics, 57, 175–195.
Gun̈gor̈, A. (2006). Evaluation of connection types in design for
disassembly (DFD) using analytic network process. Computers &
Industrial Engineering, 50(1/2), 35–54.
Hall, B. H., Jaffe, A. B., & Trajtenberg, M. (2001). The NBER patent
citations data file: Lessons, insights and methodological tools.
National Bureau of Economic Research, Working Paper No. 8498.
Ham, R. M., Linden, G., & Appleyard, M. M. (1998). The evolving role of
semiconductor consortia in the United States and Japan. California
Management Review, 41, 137–163.
Han, Y., & Park, Y. (2006). Patent network analysis of inter-industrial
knowledge flows: The case of Korea between traditional and emerging
industries. World Patent Information, 28(3), 235–247.
Harhoff, D., Scherer, F. M., & Vopel, K. (2003). Citations, family size,
opposition and the value of patent rights. Research Policy, 32(8),
1343–1363.
Hirschey, M., & Richardson, V. J. (2001). Valuation effects of patent
quality: A comparison for Japanese and US firms. Pacific-Basin
Finance Journal, 9(1), 65–82.
Jaffe, A., & Trajtenberg, M. (1998). International knowledge flows:
Evidence from patent citations. National Bureau of Economic
Research, Working Paper No. 6509.
Jharkharia, S., & Shankar, R. (2007). Selection of logistics service
provider: An analytic network process (ANP) approach. Omega, 35(3),
274–289.
Kahraman, C., Ertay, T., & Büyüközkan, G. (2006). A fuzzy optimization
model for QFD planning process using analytic network approach.
European Journal of Operational Research, 171(2), 390–411.
Karsak, E. E., Sozer, S., & Alptekin, S. E. (2003). Product planning in
quality function deployment using a combined analytic network
process and goal programming approach. Computers & Industrial
Engineering, 44(1), 171–190.
Kim, Y. G., Suh, J. H., & Park, S. C. (2007). Visualization of patent
analysis for emerging technology. Expert Systems with Applications,
doi:10.1016/j.eswa.2007.01.033.
Lai, K. K., & Wu, S. J. (2005). Using the patent co-citation approach to
establish a new patent classification system. Information Processing &
Management, 41(2), 313–330.
Lanjouw, J. O., & Schankerman, M. (1999). The quality of ideas:
Measuring innovation with multiple indicators. National Bureau of
Economic Research, Working Paper No. 7345.
Lee, J. W., & Kim, S. H. (2000). Using analytic network process and goal
programming for interdependent information system project selection.
Computers & Operations Research, 27(4), 367–382.
Lin, Y. H., Chiu, C. C., & Tsai, C. H. (2007). The study of applying ANP
model to assess dispatching rules for wafer fabrication. Expert Systems
with Applications, doi:10.1016/j.eswa.2007.02.033.
Marseden, P., & Laumann, E. (1984). Mathematical ideas in social
structural analysis. Journal of Mathematical Sociology, 10, 271–294.
Meade, L., & Presley, A. (2002). R&D project selection using ANP. IEEE
Transactions on Engineering Management, 49(1), 22–28.
Meade, L., & Sarkis, J. (1998). Strategic analysis of logistics and supply
chain management systems using the analytical network process.
Transportation Research Part E: Logistics and Transportation Review,
34(3), 201–215.
Meade, L., & Sarkis, J. (1999). Analyzing organizational project alterna-
tives for agile manufacturing processes: An analytic network approach.
International Journal of Production Research, 37(2), 241–261.
Mowery, D. C., Oxley, J. E., & Silverman, B. S. (1998). Technological
overlap and interfirm cooperation: Implications for the resource-based
view of the firm. Research Policy, 27, 507–523.
Mulebeke, J. A. W., & Zheng, L. (2006). Analytical network process for
software selection in product development: A case study. Journal of
Engineering and Technology Management, 23(4), 337–352.
Narin, F. (1994). Patent bibliometrics. Scientometrics, 30, 147–155.
Narin, F., Noma, E., & Perry, R. (1987). Patents as indicators of
corporate technological strength. Research Policy, 16, 143–155.
Niemira, M. P., & Saaty, T. L. (2004). An Analytic Network Process
model for financial-crisis forecasting. International Journal of Fore-
casting, 20(4), 573–587.
OECD (2005). Measuring globalization: OECD handbook on economic
globalization indicators. OECD, Paris.
Reitzig, M. (2004). Improving patent valuations for management pur-
poses—validating new indicators by analyzing application rationales.
Research Policy, 33(6/7), 939–957.
Ruhnau, B. (2000). Eigenvector-centrality — a node-centrality? Social
Networks, 22(4), 357–365.
Saaty, T. (1980). The analytic hierarchy process. New York: McGraw-Hill.
Saaty, T. (1996). Decision making with dependence and feedback: The
analytic network process. Pittsburgh: RWS Publications.
Schapper, M. (2003). A proposal for a core list of indicators for ICT
measurement. Electronic document at http://www.oecd.org, OECD.
Shin, J., & Park, Y. (2007). Building the national ICT frontier: The case of
Korea. Information Economics and Policy, 19(2), 249–277.
Trajtenberg, M. (1990). A penny for your quotes: Patent citations and the
value of inventions. Rand Journal of Economics, 21, 172–187.
Trajtenberg, M., Henderson, R., & Jaffe, A. B. (1997). University versus
corporate patents: A window on the basicness of invention. Economics
of Innovation and new technology, 5(1), 19–50.
USPTO (2006). Overview of the US patent classification system (USPC).
Electronic document at http://www.uspto.gov, USPTO.
Wartburg, I., Teichert, T., & Rost, K. (2005). Inventive progress measured
by multi-stage patent citation analysis. Research Policy, 34, 1591–1607.
Wasserman, S., & Faust, K. (1994). Social network analysis. Cambridge
University Press.
Wei, W. L., & Chang, W. C. (2007). Analytic network process-based
model for selecting an optimal product design solution with zero-one
goal programming. Journal of Engineering Design, doi:10.1080/
09544820601186054.
Wu, W. W., & Lee, Y. T. (2007). Selecting knowledge management
strategies by using the analytic network process. Expert Systems with
Applications, 32(3), 841–847.
Yoon, B., & Park, Y. (2004). A text-mining based patent network:
Analytical tool for high-technology trend. Journal of High Technology,
15(1), 37–50.
_ & Dag66deviren, M. (2007). Using the analytic network process
Yüksel, I.,
(ANP) in a SWOT analysis – A case study for a textile firm.
Information Sciences, 177(16), 3364–3382.