Professional Documents
Culture Documents
Download full chapter Cloud Computing Smart Grid And Innovative Frontiers In Telecommunications 9Th Eai International Conference Cloudcomp 2019 And 4Th Eai International Conference Smartgift 2019 Beijing China December 4 5 pdf docx
Download full chapter Cloud Computing Smart Grid And Innovative Frontiers In Telecommunications 9Th Eai International Conference Cloudcomp 2019 And 4Th Eai International Conference Smartgift 2019 Beijing China December 4 5 pdf docx
Innovative Frontiers in
Telecommunications 9th EAI
International Conference CloudComp
2019 and 4th EAI International
Conference SmartGIFT 2019 Beijing
China December 4 5 2019 and
December 21 22 2019 Xuyun Zhang
Visit to download the full and correct content document:
https://textbookfull.com/product/cloud-computing-smart-grid-and-innovative-frontiers-i
n-telecommunications-9th-eai-international-conference-cloudcomp-2019-and-4th-eai-i
nternational-conference-smartgift-2019-beijing-china-december-4-5/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
https://textbookfull.com/product/advanced-hybrid-information-
processing-third-eai-international-conference-adhip-2019-nanjing-
china-september-21-22-2019-proceedings-part-i-guan-gui/
https://textbookfull.com/product/advanced-hybrid-information-
processing-third-eai-international-conference-adhip-2019-nanjing-
china-september-21-22-2019-proceedings-part-ii-guan-gui/
https://textbookfull.com/product/innovative-technologies-and-
learning-second-international-conference-icitl-2019-tromso-
norway-december-2-5-2019-proceedings-lisbet-ronningsbakk/
Science and Technologies for Smart Cities 5th EAI
International Summit SmartCity360 Braga Portugal
December 4 6 2019 Proceedings Henrique Santos
https://textbookfull.com/product/science-and-technologies-for-
smart-cities-5th-eai-international-summit-smartcity360-braga-
portugal-december-4-6-2019-proceedings-henrique-santos/
https://textbookfull.com/product/broadband-communications-
networks-and-systems-10th-eai-international-conference-
broadnets-2019-xian-china-october-27-28-2019-proceedings-
qingshan-li/
https://textbookfull.com/product/cloud-computing-and-services-
science-9th-international-conference-closer-2019-heraklion-crete-
greece-may-2-4-2019-revised-selected-papers-donald-ferguson/
https://textbookfull.com/product/e-learning-e-education-and-
online-training-5th-eai-international-conference-
eleot-2019-kunming-china-august-18-19-2019-proceedings-guanglu-
Xuyun Zhang
Guanfeng Liu
Meikang Qiu
Wei Xiang
Tao Huang (Eds.)
322
123
Lecture Notes of the Institute
for Computer Sciences, Social Informatics
and Telecommunications Engineering 322
123
Editors
Xuyun Zhang Guanfeng Liu
Macquarie University Macquarie University
Sydney, NSW, Australia Sydney, NSW, Australia
Meikang Qiu Wei Xiang
Pace University James Cook University
New York, NY, USA Cairns, QLD, Australia
Tao Huang
James Cook University
Smithfield, QLD, Australia
© ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2020
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the
material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now
known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book are
believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors
give a warranty, express or implied, with respect to the material contained herein or for any errors or
omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
We are delighted to introduce the proceedings the 9th edition of the European Alliance
for Innovation (EAI) International Conference on Cloud Computing (CloudComp
2019). This conference brought researchers, developers, and practitioners from around
the world together who are leveraging and developing cloud computing related tech-
nology for IoT and big data applications.
The technical program of CloudComp 2019 consisted of 48 full papers from several
tracks. The conference tracks were: Track 1 – Cloud Architecture and Scheduling;
Track 2 – Cloud-based Data Analytics; Track 3 – Cloud Applications; and
Track 4 – Cloud Security and Privacy. Aside from the high-quality technical paper
presentations, the technical program also featured three keynote speeches and a
Distinguished Local Scholar Discussion Panel. The three keynote speeches were
delivered by Prof. Michael Sheng from Macquarie University, Australia, Prof.
Yew-soon Ong from Nanyang Technological University, Singapore, and Prof.
Longbing Cao from University of Technology Sydney, Australia. The scholars who sat
on the Distinguished Local Scholar Discussion Panel were Principle Scientist Surya
Nepal from CSIRO’s Data61, Australia, Prof. Shui Yu from University of Technology
Sydney, Australia, Dr. Lina Yao from University of New South Wales, Australia, Dr.
Qiang He from Swinburne University of Technology, Australia, Dr. Dong Yuan from
The University of Sydney, Australia, and Dr. Jia Wu from Macquarie University,
Australia. The panel discussion aimed to gain insights into key challenges and an
understanding of the state-of-the-art cloud computing technologies to develop and
implement future cloud and edge computing related services and applications.
Coordination with the Steering Committee chairs, Imrich Chlamtac, Victor
C. M. Leung, and Kun Yang was essential for the success of the conference. We
sincerely appreciate their constant support and guidance. It was also a great pleasure to
work with such an excellent Organizing Committee, and we thank them for their hard
work in organizing and supporting the conference. In particular, we thank the Technical
Program Committee (TPC), led by our TPC co-chairs, Dr. Deepak Puthal, Dr. Chi
Yang, and Dr. Zhiyuan Tan, who organized the peer-review process of technical papers
and made a high-quality technical program. We are also grateful to the conference
manager, Karolina Marcinova, for her support and to all the authors who submitted
their papers to the CloudComp 2019 conference.
We strongly believe that the CloudComp 2019 conference provides a good forum
for all researchers, developers, and practitioners to discuss all science and technology
aspects that are relevant to cloud and edge computing. We also expect that the future
CloudComp conferences will be as successful and stimulating, as indicated by the
contributions presented in this volume.
We are delighted to introduce the proceedings of the second edition of the European
Alliance for Innovation (EAI) International Conference on Smart Grid Inspired Future
Technologies (SmartGIFT 2019). This conference brought researchers, developers, and
practitioners from around the world together who are leveraging and developing smart
grid technology for a smarter and more resilient grid. The theme of SmartGIFT 2019
was “Smart Grid and Innovative Frontiers in Telecommunications”.
The technical program of SmartGIFT 2019 consisted of 14 full papers, including 3
invited papers in oral presentation sessions at the main conference track. Aside from the
high-quality technical paper presentations, the technical program also featured a key-
note speech. The keynote speech was presented by Prof. Zhaocheng Wang from
Tsinghua University, Beijing, China. This keynote speech was focused on the tech-
nology of AI-based low-power millimeter wave communications.
Coordination with the Steering Committee chairs, Imrich Chlamtac, Victor
C. M. Leung, and Kun Yang was essential for the success of the conference. We
sincerely appreciate their constant support and guidance. It was also a great pleasure to
work with such an excellent Organizing Committee, and we thank them for their hard
work in organizing and supporting the conference. In particular, we thank the Technical
Program Committee (TPC), led by our TPC chair, Dr Tao Huang, who completed the
peer-review process of technical papers and made a high-quality technical program. We
are also grateful to the conference manager, Lukas Skolek, for his support and to all the
authors who submitted their papers to the SmartGIFT 2019 conference and workshops.
We strongly believe that the SmartGIFT conference provides a good forum for all
researcher, developers, and practitioners to discuss all science and technology aspects
that are relevant to smart grids. We also expect that the future SmartGIFT conference
will be as successful and stimulating, as indicated by the contributions presented in this
volume.
Steering Committee
Imrich Chlamtac University of Trento, Italy
Organizing Committee
General Chair
Xuyun Zhang Macquarie University, Australia
General Co-chairs
Xuyun Zhang Macquarie University, Australia
Meikang Qiu Pace University, USA
Local Chair
Dong Yuan The University of Sydney, Australia
Workshops Chair
Lianyong Qi Qufu Normal University, China
Publications Chairs
Shaohua Wan Zhongnan University of Economics and Law, China
Xiaolong Xu Nanjing University of Information Science and
Technology, China
x Organization CloudComp 2019
Web Chair
Meng Liu Shandong University, China
Conference Manager
Karolina Marcinova EAI
Steering Committee
Imrich Chlamtac University of Trento, Italy
Victor C. M. Leung The University of British Columbia, Canada
Kun Yang University of Essex, UK
Organizing Committee
General Chair
Wei Xiang James Cook University, Australia
General Co-chair
Lin Cao Beijing Information Science and Technology
University, China
TPC Chair
Tao Huang James Cook University, Australia
Local Chair
Xuehua Li Beijing Information Science and Technology
University, China
Publications Chair
Yuanyuan Yao Beijing Information Science and Technology
University, China
Panels Chair
Yanxiao Zhao Beijing Information Science and Technology
University, China
Web Chair
Eric Wang James Cook University, Australia
Conference Manager
Lukáš Školek EAI
Cloud Applications
Enhanced LSTM Model for Short-Term Load Forecasting in Smart Grids . . . 650
Jianing Guo, Yuexing Peng, Qingguo Zhou, and Qingquan Lv
1 Introduction
The emerging Location-Based Social Network (LBSN) attracts a large number of partic-
ipants. Foursquare, Twitter and other applications make it easy for people to mark their
locations. This has aroused widespread interests among researchers. In LBSN, Multi-
Constrained Optimized Path Selection (MCOPS) has become a challenging problem. In
this paper, we study a Spatial Pattern Matching (SPM) problem based on MCOPS, that
is, in the LBSN, to find all the matches satisfying multi-constrained path pattern P in
the space objects set D. The multi-constraints in P include keyword constraint, distance
constraint, and social impact factors (trust, intimacy, social reputation) constraints. The
related Spatial Keyword Query (SKQ) and Graph Pattern Matching (GPM) have been
extensively studied.
In general, a spatial-keyword query (SKQ) [1–4] returns spatial objects that are spa-
tially close to each other and match the keywords. For example, top-k keyword query [5],
takes the spatial location and a set of keywords as parameters, returning k spatio-textual
© ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2020
Published by Springer Nature Switzerland AG 2020. All Rights Reserved
X. Zhang et al. (Eds.): CloudComp 2019/SmartGift 2019, LNICST 322, pp. 3–19, 2020.
https://doi.org/10.1007/978-3-030-48513-9_1
4 Y. Guo et al.
objects that well meet the user’s requirements, and the k objects are sorted according to
the proximity of the query location and the relevance of the query keywords. Although
SKQ plays an important role in geo-location-oriented applications (e.g., Google Maps),
however, with the continuous emergence of new application scenarios, the requirements
of users are increasingly diversified, which results in the returned query results not
meeting the needs of users [6]. Moreover, various relationships between social network
participants (e.g., location relationships, social relationships) are not well represented.
Suppose that a user wishes to find a supplier, which is close to a bank and is far from
a retailer and a whole-seller. SKQ will return objects that are spatially close to each
other as shown in the dashed box in Fig. 1(a). However, users generally have restrictions
on the spatial objects of the query (e.g., the supplier is at least 10 km away from the
retailer in Fig. 1(b)). Therefore, the objects connected by the solid line in Fig. 1(a) are
the answer to the query that the user really wants. Second, simply using spatial indexing
structures (such as R-tree and their variants) requires dynamically creating an index for
each vertex, which results in serious overhead [7]. Therefore, the solution to the SKQ
problem cannot be directly applied to the MCOPS-SPM problem.
bank
20 whole-seller
{bank,supermarket} dis=[0,2]
supplier retailer td>=0.7
15 si>=0.1
doctor economist supplier
sr>=0.5
10 dis>=10 dis=[10,20]
td>=0.6 td>=0.7
si>=0.05 si>=0.2
5 whole-seller
retailer sr>=0.4 sr>=0.5
In general, graph pattern matching (GPM) aims to find all matches of P in data
graph G according to a given pattern graph P [8]. However, using GPM algorithms
to solve MCOPS-SPM problem is not straightforward because (1) the solution to the
GPM problem is primarily designed for the data structure of the graph, rather than spatio-
textual objects indexed by an IR-tree index structure, (2) in the pattern graph, only a single
constraint (such as hops or distances) among vertices is generally considered. Therefore,
if GPM is used to solve the MCOPS-SPM problem, it is necessary to convert the spatial
objects into graphs and extend the distance-based constraint to multi-constraints. The
experiments in [7] prove that it is not efficient.
There are few studies on Spatial Pattern Matching (SPM). [9] proposed a database
query language based on time mode, spatial mode and spatiotemporal mode. In [7], a
SPM algorithm is proposed, but only a single constraint (different expression of distance
constraint) is considered, which cannot meet the requirements of the user’s multiple
constraints.
We use an example in [10] to illustrate the importance of specifying multiple con-
straints for query keywords. In business activities, the pattern graph can be specified to
look for keywords about suppliers, retailers, whole-sellers, banks in the social network.
MCOPS-SPM: Multi-Constrained Optimized Path Selection Based SPM 5
The supplier directly or indirectly provides the retailer, whole-seller with products. Sup-
pliers, retailers, whole-sellers obtain services directly or indirectly from the same bank.
As shown in Fig. 1, the user specifies the supplier is at least 10 km from the retailer and
the trust value is at least 0.6, the intimacy is at least 0.05, and the social reputation is at
least 0.4 in the field. However, SKQ and GPM tend to ignore the distance constraint and
social impact factor constraints of these problems, and the results of the query are often
not of the most interest to the users.
In order to facilitate users to specify the spatial relationships between the keywords
and the social relationships, we propose a multi-constrained optimized path selection
based spatial pattern matching query (MCOPS-SPM). As shown in Fig. 1, given a set of
spatial objects D (Fig. 1 (a)) and a spatial path pattern P (Fig. 1 (b)), MCOPS-SPM returns
all matches about P in D. The pattern P is a complex network in which vertices represent
spatial objects associated with keywords, edges represent spatial distance relationships
between objects, and social relationships. For example, the distance between the user-
specified bank and the supplier is [0, 2] (km), the trust value is at least 0.7, the intimacy
is at least 0.1, and the social reputation is at least 0.5. In this example, the four objects
connected by solid lines satisfy all constraints in P, which is a match of pattern P.
The main research contents of this paper are as follows:
The rest of this paper is organized as follows. We first review the related work on
SKQ and GPM in Sect. 2. Then we introduce the necessary concepts and formulate the
focal problem of this paper in Sect. 3. Section 4 presents our solutions MCOPM, IR-
MCOPM, MCPJoin algorithms. We report experimental results in Sect. 5 and conclude
in Sect. 6.
6 Y. Guo et al.
2 Related Works
The most related works about spatial pattern matching are spatial keyword query and
graph pattern matching. Below we analyze each of them in detail.
still very harsh for applications that need to meet the specified path length. In order to
solve the above problem, a bounded simulation is proposed in [8], in which each node
has an attribute, each edge is associated with the path length (hops), and the value of
each edge is One of len and *, len indicates that the path length cannot exceed len, and
* indicates that there is no constraint on the path length. The bounded simulation maps
the edges in pattern P to paths that meet the specified length in the data graph, further
reducing the strict constraints, thereby better capturing users’ intents.
There are some algorithms that connect the searched edges in a certain order [21].
In [21], an MDJ (Multi-Distance-Join) algorithm is proposed to guide the connection
order of edges. In [10], the author proposed an R-join algorithm based on reachability
conditions, using the B+ -tree structure as the index method.
However, the solutions above mentioned are based on a single constraint algorithm,
and can not meet the requirements of multiple constraints specified by users, so these
methods can not be directly applied to the MCOPS-SPM problem.
3 Problem Definition
3.1 Social Impact Factors
Social networks can be modeled as directed graphs, and we use G = (V, E) to express
social networks. Where V = {v1 , v2 , . . . , vn } represents a collection of vertices, each
vertex vi represents a participant in a social network. E = {e1 , e2 , . . . , em } is a collection
representing edges, each edge ej represents an interaction or social relationships between
two participants. However, the factors that influence social networks are always diverse,
and we cannot draw precise conclusions based on a single condition. We propose four
factors that influence social networks:
Social Identities. In a social network, each participant has his or her own identities, rep-
resented by keywords k(vi ). Each participant may have multiple identities (for example,
a digital blogger may be a student at a university), so the social identities help describe
the social roles of participants.
Trust Degree. In a social network, trust degree refers to the level of trust that one par-
ticipant forms in the interaction process with another participant. Let vi , vj ∈ [0, 1]
td
indicate the trust degree that participant
v
i evaluates v j . If td vi , v j = 1, it represents
that vi completely trusts vj while td vi , vj = 0 represents vi wholly distrusts vj .
Social Intimacy. In social networks, intimacy refers to the degree of intimacy formed
between participants during the participation of social
activities.
It reflects the frequency
of interaction between the two participants.
si v ,
i jv ∈ [0, 1] reflects the degree of
intimacy between vi and vj . When si vi , vj = 1, it expresses that vi and vj are the most
intimate. On the contrary, they have the least intimate social relationships.
Social Reputation. In social networks, social reputation refers to the extent to which a
participant contributes in the field. It reflects the extent to which all participants rated this
participant. Let sr(vi ) ∈ [0, 1] denote social reputations in the domain. Here sr(vi ) = 1
represents that vi has won the best reputations in the domain. Otherwise, there is no
foothold for vi in this field.
8 Y. Guo et al.
For example, in the pattern P of Fig. 1(b), the limit of distance between bank and
supplier is at [0, 2] km, trust degree is no less than 0.7, social intimacy is not less than
0.1, social reputation is no less than 0.5.
Another random document with
no related content on Scribd:
city, membership in which now, I was informed, amounted to
10,000. The chief organizer was a shrewd mestizo, a former
close companion of Aguinaldo, by whom he had been commissioned
to perform this work. He was a friend and associate of some of
our officers; was engaged in organizing the clubs only, as he
stated, to give the poorer classes amusement and education;
held public entertainments in athletics to which our officers
were invited, and in which our soldiers were asked to
participate. Gradually arms were being secretly introduced and
bolos were being manufactured and distributed. The arms were
kept concealed in buildings, and many of them were
subsequently captured. The Chinamen were carrying on a
lucrative business in bolo making, but the provost-marshal had
cruelly seized considerable of their stock. These clubs had
received military organization and were commanded by cunning
Filipino officers regularly appointed by the Malolos
government. The chief organizer departed after organization
had been perfected and thereafter became a confidential
adviser in Malolos affairs. This organization was the subject
of grave apprehension, as it was composed of the worst social
element of the city, and was kept under police supervision as
closely as possible. … The streets of the city were thronged
with unarmed insurgent officers and enlisted men from the
numerically increasing insurgent line on the outskirts, proud
of their uniforms and exhibiting matchless conceit, amusing to
our men, who were apparently unconcerned observers, but who
were quick to take in the rapidly changing conditions. …
WILLIAM McKINLEY."
{379}
"No definite date had been set for the attack [by the hostile
Tagalos], but a signal by means of rockets had been agreed
upon, and it was universally understood that it would come
upon the occurrence of the first act on the part of the
American forces which would afford a pretext; and in the lack
of such act, in the near future at all events. Persistent
attempts were made to provoke our soldiers to fire. The
insurgents were insolent to our guards and made persistent and
continuous efforts to push them back and advance the insurgent
lines farther into the city of Manila. … With great tact and
patience the commanding general had held his forces in check,
and he now made a final effort to preserve the peace by
appointing a commission to meet a similar body appointed by
Aguinaldo and to 'confer with regard to the situation of
affairs and to arrive at a mutual understanding of the intent,
purposes, aims, and desires of the Filipino people and of the
people of the United States.' Six sessions were held, the last
occurring on January 29, six days before the outbreak of
hostilities. No substantial results were obtained, the
Filipino commissioners being either unable or unwilling to
give any definite statements of the 'intent, purposes, and
aims of their people.' At the close of the last session they
were given full assurances that no hostile act would be
inaugurated by the United States troops. The critical moment
had now arrived. Aguinaldo secretly ordered the Filipinos who
were friendly to him to seek refuge outside the city. The
Nebraska regiment at that time was in camp on the east line at
Santa Mesa, and was guarding its front. For days before the
memorable 4th of February, 1899, the outposts in front of the
regiment had been openly menaced and assaulted by insurgent
soldiers; they were attempting to push our outposts back and
advance their line. They made light of our sentinels and
persistently ignored their orders. On the evening of the 4th
of February, an insurgent officer came to the front with a
detail of men and attempted to pass the guard on the San Juan
Bridge, our guard being stationed at the west end of the
bridge. The Nebraska sentinel drove them back without firing,
but a few minutes before 9 o'clock that evening a large body
of insurgent troops made an advance on the South Dakota
outposts, which fell back rather than fire. About the same
time the insurgents came in force to the east end of the San
Juan Bridge, in front of the Nebraska regiment. For several
nights prior thereto a lieutenant in the insurgent army had
been coming regularly to our outpost No. 2, of the Nebraska
regiment, and attempting to force the outpost back and
insisting on posting his guard within the Nebraska lines; and
at this time and in the darkness he again appeared with a
detail of about six men and approached Private Grayson, of
Company D, First Nebraska Volunteers, the sentinel on duty at
outpost No. 2. He, after halting them three times without
effect, fired, killing the lieutenant, whose men returned the
fire and then retreated. Immediately rockets were sent up by
the Filipinos, and they commenced firing all along the line, …
and continued to fire until about midnight; and about 4
o'clock on the morning of February 5 the insurgents again
opened fire all around the city and kept it up until the
Americans charged them and drove them with great slaughter out
of their trenches."
{381}
"By the 10th of October the process of changing armies and the
approach of the dry season had reached a point where an
advance toward the general occupation of the country was
justified. At that time the American lines extended from the
Bay of Manila to Laguna de Bay, and included considerable
parts of the provinces of Cavite, Laguna, and Morong to the
south and east of Manila, substantially all of the province of
Manila and the southern parts of Bulacan and Pampanga,
dividing the insurgent forces into two widely separated parts.
To the south and east of our lines in Cavite and Morong were
numerous bands occasionally concentrating for attack on our
lines, and as frequently dispersed and driven back toward the
mountains. On the 8th or October, the insurgents in this
region having again gathered and attacked our lines of
communication, General Schwan with a column of 1,726 men
commenced a movement from Bacoor, in the province of Cavite,
driving the enemy through Old Cavite, Noveleta, Santa Cruz,
San Francisco de Malabon, Saban, and Perez das Marinas,
punishing them severely, scattering them and destroying them
as organized forces, and returning on the 13th to Bacoor. On
the north of our lines stretched the great plain of central
Luzon extending north from Manila about 120 miles. This plain
comprises parts of the provinces of Manila, Pampanga, Bulacan,
Tarlac, Nueva, Ecija, and Pangasinan. It is, roughly speaking,
bounded on the south by the Bay of Manila: on the east and
west by high mountain ranges separating it from the seacoasts,
and on the north by mountains and the Gulf of Lingayen.
Through the northeast and central portion flows the Rio Grande
from the northern mountains southwesterly to the Bay of Manila,
and near the western edge runs the only railroad on the island
of Luzon, in a general southeasterly direction from Dagupan,
on the Bay of Lingayen, to Manila. In this territory Aguinaldo
exercised a military dictatorship, and with a so-called
cabinet imitated the forms of civil government, having his
headquarters at Tarlac, which he called his capital, and which
is situated near the center of the western boundary of the
plain.
{382}