Dynamic Generation For System Over Database

International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org

Volume 3, Issue 12, December 2014
ISSN 2319 - 4847
Dynamic generation for system over database

A.Mallareddy1, J.Niharika2 and Ch. Vasavi3
1
Research Scholar(JNTUH), Department of Computer Science & Engineering

Professor &HOD(CSE) Sri Indu Institute of Engineering & Technology
Sheriguda(V), Ibrahimpatnam(M), RR Dist 501510
2
M.Tech (CS), Department of Computer Science & Engineering
Sri Indu Institute of Engineering & Technology, Sheriguda(V), Ibrahimpatnam(M), RR Dist 501510
3
Associate Professor, Department of Computer Science & Engineering
Sri Indu Institute of Engineering & Technology, Sheriguda(V), Ibrahimpatnam(M), RR Dist 501510
ABSTRACT
Relation knowledge-base systems are becoming increasingly pleasing to all in the scientific community to support the effecting
on one another discovery of greatly sized volumes of facts. In this scenario, Users use a question connection (representatively,
a web-based client) to question under discussion a number, order, group, line of SQL questions that try to get at the details of
the knowledge for computers and mine it for interesting information. First-time Users, however, may not have the necessary
knowledge to have knowledge of where to start their discovery. Other times, Users may simply overlook questions that get back
important information. To help Users in this makes sense clearer, we outline inspiration from net of an insect recommender
systems and make an offer the use of made for a person question statements of good words for. The idea is to unbroken bands
over wheels for moving over rough earth the questioning behavior of each User, make out which parts of the knowledge-base
may be of interest for the being like (in some way) knowledge for computers observations work, and suggest questions that get
back on the point facts. We have a discussion the main questions in this fiction story use of recommendation systems, and
outline a possible answer based on collaborative coming through slowly. Preliminary testing results on true User outlines put
examples on view that our framework can produce working well question statements of good words for.
1 INTRODUCTION
Query form is one of the most widely used User connections for questioning knowledge-bases old and wise question,
forms are designed and selected before by ones that makes or DBA in different information managers of business
systems. With the quick development of net of an insect information and scientific knowledge-bases of-the-day
knowledge-bases become very greatly sized and complex. In natural sciences such as genomics and diseases the
knowledge-bases have over hundreds of things for chemical and biological data resources. Many net of an insect
knowledge-bases such as Freebase and DBPedia representatively have thousands of structured net of an insect things as
an outcome of that it is hard to design a put of at rest question forms to give what is desired, needed to different after
Christ hoc knowledge-base questions on those complex knowledge-bases. Relation knowledge-base systems are
becoming increasingly pleasing to all in the scientific community to manage greatly sized volumes of based on
experience facts. Examples join the genome browser that provides way in to a genomic knowledge-base, and Sky Server
that stores greatly sized volumes of great measurements. The main better chances of an of relation knowledge-base
system is that it supports the good at producing an effect wrongdoer put to death of complex questions, thus making
able Users to effecting one another have a look for the facts and get back interesting information. It should be noted
that the named before systems use web-based question connections in order to be able be got to a wide User base. Even
though a knowledge-base system offers the means to run complex questions over greatly sized facts puts, the discovery
of useful information remains a great-sized sporting offer. Users who are not every day with the knowledge-base may
overlook questions that get back interesting facts, or they may not have knowledge of what parts of the knowledge-base
make ready useful information. This question under discussion clearly gets in the way of facts discovery, and thus gets
changed to other form the benefits of using a knowledge-base system. To house this important hard question, we
outline inspiration from the good use of recommender systems in the discovery of net of an insect fact. In particular, we
chief place on moves near based on user-based collaborative coming through slowly. The statement on which reasoning
is based is simple: If a User A has similar questioning behavior to User b, then they are likely interested in the same
facts. For this reason, the questions of User b can put ball in play as a guide for User A. The get moved from one
position to another of this example gives property in line several hard questions. In net of an insect collaborative
coming through slowly systems, and user-item matrix move near is used to produce statements of good words for. More
specially, each User is represented as an one thing on a list guide, where the values of the guide elements be like to the
Users desires for each one thing on a list (such as motion picture ratings, got to own products, read articles, and so on.)
The similarities between Users in this pictures of can be easily worked out using guide similarity metrics. given the
Page 56

ISSN 2319 - 4847
most similar Users and their desires, the collaborative coming through slowly system can coming after say what will
take place in the future what items will interest each User, and produce one thing on a list statements of good words
for. The named before methodology cannot be directly sent in name for to the makes sense clearer of an of relation
knowledge-base for several reasons. First, we observe that in the example of a knowledge-base the items of interest are
knowledge-base records, and the Users way in these items in a roundabout way by taking a position SQL questions. In
this way, even though the Users behavior is taken to be by the group of questions they send to the knowledge-base, their
interest lies on the knowledge-base tuples they put in good order again. Given that SQL is a declarative language,
however, the same facts can be got back in more than one way. This makes complex the put value of similarity among
Users based on their questions by oneself, since it is no longer clearly and readily seen whether they are interested in
the same things on a list. This raises a second important question under discussion that needs some thought. The
similarity between Users can be expressed as the similarity between the parts of their questions or, in a different way,
the facts that they get back. This is not as straightforward, since a question part or a tuple might have different levels of
importance in different User meeting. In this way, we must be able to make come into existence if true, then some other
is necessarily true User face seen from the side that design to be copied those levels of importance, in order to
effectively make a comparison the Users. At last, opposite to the user-based collaborative coming through slowly move
near, the recommendation to the Users have to be in the form of SQL questions, since those actually make, be moving
in what they got back facts represent. in this way, we need to close the do orders till a condition by first decomposing
the User questions into lower-level parts in order to work out similarities and make statements of what will take place
in the future, and then re-construct them back to SQL questions in order to suggest them. in addition, these SQL
questions must be purposeful and intuitive, so that Users can take into parts them and get clearly their having attention
fixed (on) and usefulness. All those issues make the hard question of effecting on one another knowledge-base
discovery very different from its net of an insect thing like another.
1.1 Our Approach
In this paper we make an offer a forcefull question Form system DQF a question connection which is able of with
motion producing question forms for Users. Different from old and wise printed material acts to get back Users in
knowledge-base acts to get back are often ready to act many rounds of actions i.e. making clean question conditions
before making out the last candidates. The liquid with special qualities of DQF is to take User interests during User
effects on one another and to adjust the question form again and again Each iteration is chiefly of two types of User
effects on one another question Form Enrichment and question wrongdoer put to death see Table 1 number in sign 1
shows the work moving liquid of DQF It starts with a Basic question form which has in it very few first properties of
the knowledge-base The Basic question form is then made better off again and again via the effects on one another
between the User and our system until the User is pleased with the question results In this paper we mainly work-room
the position on scale of question form parts and the forcefull stage of question forms. Our contributions can be made a
short account as follows We make an offer a forcefull question form system which produces the question forms
according to the User s desire at run time The system provides an answer for the question connection in greatly sized
and complex knowledge-bases We send in name for F measure to value the goodness of a question form F measure is
an of a certain sort metric to value question results This metric is also right for question forms because question forms
are designed to help Users question the knowledge-base The goodness of a question form is strong of purpose by the
question results produced from the question form based on this we degree and suggest the possible & unused quality
question form parts so that Users can make clean the question form easily based on the made an offer metric we
undergo growth good at producing an effect Algorithms to value the goodness of the thing coming out from and
selection form parts Here doing work well is important because DQF is a connected system where Users often hope for
quick move.
TABLE 1 Interactions between User and DQF
Page 57

ISSN 2319 - 4847
2 Related Work
So far, the work that has been done in the area of made for a person knowledge-base has gave one's mind to an idea to
keyword-based question recommendation systems. In this scenario, a User can acts between, along with an of relation
knowledge-base through a net of an insect connection that lets him/her to take orders (from) keywords and get back on
the point What is in. The personalization process is based on the Users keyword questions, those of earlier Users, as
well as a clear and detailed User outline that records the Users desires with in connection with to the What is in of the
knowledge-base. Clearly, our move near is different from this scenario in several ways. First, the made an offer
framework is meant to help Users who take up a position complex SQL questions to of relation knowledge-bases. In
addition, the system does not have need of from its Users to make come into existence a clear and detailed outline. This
gives a higher level of able to make ready adjustments to the system, since the same User might have different
information needs during different test of the knowledge-base. Our inspiration pulls up from the good use of userbased collaborative coming through slowly expert ways of art and so on, made an offer in the Web makes sense clearer.
As previously said-about, this move near cannot be directly sent in name for to the of relation knowledge-base makes
sense clearer. The natural to nature of effecting on one another knowledge-base discovery raises certain follow up that
cannot be made house numbers by the straightforward collaborative coming through slowly move near. In this work, we
are based on its in the building, but give (kind attention) them in order to send in name for them in the knowledgebase general condition. The questions of putting to use facts mining techniques to the knowledge-base question records
are also made house numbers in. In this work, the writers outline the buildings and structure design of a Collaborative
question managers of a business System marked free-scale, shared-data conditions. As part of this buildings and
structure design, they not dependently suggest that facts mining expert ways of art and so on, such as clustering or
organization rules, can be sent in name for to the question records in order to make ready the Users with question
suggestions. We should conditions making things hard, however, that opposite to our work, the writers do not make
ready any special to some science or trade details on how such a recommendation system could be gave effect to.
3 PERSONALIZED QUERY RECOMMENDATIONS

The hard question of made for a person question recommendations can be put clearly as takes as guide, example, rule:
given a User that is currently exploring the knowledge-base, suggest questions that might be of interest to him/her. To
produce such statements of good words for, the system will get support from on information gathered from the
questioning behavior of past Users, as well as the questions put forward by the current User so far. The information
move liquid-like of the QueRIE framework is made clear in Figure1. The action-bound Users questions are forwarded
to both the DBMS and the recommendation Engine. The DBMS processes each question and comes back a group of
comes out. At the same time, the question is stored in the question record. The recommendation Engine trading groups
the current Users input with information gathered from the knowledge-base effects on one another of past Users, as
recorded in the question record, and produces a group of question recommendations that are returned to the User.
In what takes as guide, example, rule, we make out the current User with the part of mind given to pleasure 0, and note
that Q0 has in it the questions that the User has put forward thus far. We use {1,... .,H} to be the sign of the group of
past Users based on which recommendations are produced. The supporters parts maker, be moving in an answer to
this interesting hard question of producing made for a person question statements of good words for. We begin by
having a discussion a with the idea framework that can go round different moves near, and then make an offer an
instantiation based on collaborative coming through slowly.
Page 58

ISSN 2319 - 4847
3.1 Conceptual Framework

Clearly, the questions of each User touch and a division of the knowledge-base that is on the point for the observations
the User wants to perform. We take to be true that this a division of is designed to be copied as a meetings short
account. This short account takes the parts of the knowledge-base made way in by the User and makes into company a
metric of importance for each part. For example, an un worked short account may have within the names of the
relations that come into view as in the questions of the User, and the importance of each relation can be measured as
the number of questions that statement, direction it. On the other very much, a detailed short account may have within
the current results carefully looked at by the User, in company with a clear and detailed rating of each outcome tuple,
taking to be true that the good quality of the short account is fixed for all Users, we use Si to be the sign of the short
account for User i. To produce statements of good words for, our framework produces a predicted short account S0 pred .
This short account takes the predicted degree of interest of the action-bound User with respect to all the parts of the
knowledge-base, including those that the User has not had a look for still, and thus serves as the seed for the complete
persons living time of statements of good words for. As an example, if the short account S0 has in it the names of the
relations that the User has made, used statement, direction so far, then Spred may have within more relations that might
be of interest, in company with the separate degree of interestingness for each part. Using S0pred , the framework makes
questions that cover the division of the knowledge-base with the highest predicted importance. In turn, these questions
are presented to the User as statements of good words for. Overall, our framework is chiefly of three parts: (a) the
making of a meetings short account for each User i based on the questions in Qi, (b) the computation of S0pred based on
the action-bound User S0 and the short accounts S1, . . . , Sh of past Users, and (C 9) the stage of questions based on
S0pred . An interesting point is that parts (a) and (C 9) form a shut circle, going from questions to short accounts and
back. This is a conscious design good quality supporters the fact that all User effect on one another with an of relation
knowledge-base comes to mind through declarative questions. The high-purpose metric for the put feelings power of a
connection is the fraction of User questions pleased in a representative User question record. But getting such a record
is not unimportant or every day. In fact it is not possible for a new knowledge-base put out for the first time. Thus our
sporting offer is to design a good group of forms without having a current question make record at hand based one and
only on the schema and data, What is in of a knowledge-base we must give opinion to a designer what forms to make
come into existence given the greatly sized range of questions that could possibly be put forward against any
knowledge-base this wall to keep persons out appears to be undoable how can we select a small number of questions to
support using a few simple forms from this greatly sized group of possible states. To profit an intuition for how schema
and What is in of the knowledge-base can make ready a reasonable starting point to value User question needs give
thought to a motion picture knowledge-base. It is likely to have movies as the most important and highest full with
group thing Users making way in such a knowledge-base will most likely look for information about movies as opposite
to information about producers persons making distribution directors writers acting persons and so on which are
important but less frequently question for this reason a form that gives one's mind to an idea on motion picture related
information is likely to be used in a wide ranging way and free from doubt an important fraction of User questions Of
direction this is a very very simple example A slightly more complex schema that many knowledge-base researchers are
everyday with is the XMark point of comparison. In its schema the first things are open put up for offers shut put up for
offers one thing on a list and person through an unreasoned feeling these are the things of most interest to Users whose
questions may get rolled in one or more of these things Most current knowledge-bases however have much more
complex schemas with a great number of things properties and relationships. Many of these knowledge-bases are
normalized to a greatly sized amount which increases the number of things needed to take the data of interest. In most
cases the schema being complex simply needs payment to the fullness of the data. This being complex is had thoughts
in the questions to the knowledge-base many with more than one thing of interest. In this paper we make, be moving in
our move near to come to a decision about exactly that We can break the forms connection design hard question down
into three questions had a discussion about below.
3.2 FORM COMPOSITION
Having worked out the queriabilities of things collections of related things and properties we then select the most
queriable of them to make up forms When we take into account these three types of schema parts in turn we see that
they are not Co equal they do not have a relation with to one another in a like in size way For example we will often get
it useful to question a thing even with no related things included in the form In comparison questioning a group of
related things always means questioning the forming part of things in the same way properties can be present in forms
only as forming part of parts of their father or mother things. This leads us naturally to a three step selection process to
make up forms First we select a thing and make come into existence a form for it the one after we select other things
related to it and make come into existence added forms one for each nothing like it relation and its taking part things at
last we select properties for each thing and place them in each form having in it that thing. Selecting FORM parts, We
first sort the things in droping order of queriability and select the top ke of them to make up forms ke is a pre detailed
threshold selected by the connection designer to control the size of connection produced by the system. Each of the
selected top ke things will be the question chief place of one form in the produced connection nearest we take into
thought related things. We work out the queriability of every group of things related by a schema defined relation
Page 59

ISSN 2319 - 4847
having in it at least one thing position on scale in the top ke most queriable. We then take into account each of the top
ke things degree its related thing collections by queriability signs making clear and select its top Kr collections Kr is
a threshold that serves as an upper joined on the number of forms made come into existence for related thing
collections. We then make come into existence a new form for each group In do however we only take into account
straight to based on relationships and roundabout based on relationships i.e., relationships in which the two things are
not related directly but are both related a common third thing based on relationships between things that are as if far
away related i.e., connected by a footway of length greater than two in the schema graph as well as ternary and higher
arity relationships are not thought out as because they cause considerably higher listing and computation costs without
a high chance of getting (grain) cut like benefits the queriabilities of these related thing twos are hard to say what will
take place in the future even with heuristics and making into company them can lead to forms of little interest to Users
at last we take into account the properties of each thing to place in these forms. We work out the queriability of each
property and degree them by queriability. We then select the top ka properties of each thing for (thing) taken in each
form having in it that thing ka is a pre detailed system threshold that controls the being complex of each person form In
Fig 1 we make clear to the outcome of our queriability computations for the XMark schema. The XMark knowledge
we used was produced using the point of comparison by frame for events the scaling cause to The thresholds were
group at things ke properties per thing ka and 5 related thing twos per thing Kr .
4 QUERY FORM INTERFACE

Our work is different as we are giving out playing cards with knowledge-base question forms instead of data-entry
forms. Action-bound point Probing: Zhu et Al. undergo growth the action-bound featuring probing expert way of art
and so on for automatically producing be made clear questions to make ready right recommendations to Users in
knowledge-base look for. Different from their work which gives one's mind to an idea on discovering the right
questions to question the User, DQF try to select right question parts.
4.1 Query Results
Question results to come to a decision whether a question form is desired or not, a User does not have time to go over
every facts example in the question outcomes. In addition, many knowledge-base questions output a very great amount
of facts examples. In order to keep from this Many-Answer hard question, we only output a make shorter outcome table
to play or amusement a high level view of the question results first. Each example in the make shorter table represents a
cluster of current facts examples. Then, the User can click through interested clusters to view the detailed facts
examples. Number in sign shows the move liquid-like of User acts. The make shorter high-level view of question
results is made an offer in. There are many one-pass clustering Algorithms for producing the make shorter view with
small amount of money, In our putting into effect, we select the incremental facts clustering framework because of the
doing work well question under discussion. Certainly, different facts clustering methods would have different make
shorter views for the Users. In addition, different clustering methods are preferable to different data sorts. In this
paper, clustering is just to make ready a better view of the question results for the User. The system ones that makes can
select a different clustering Algorithm if needed. Another important use of the make shorter view is to keep (self,
thoughts) in order, under control the User take-back. Using the self control get food to back, we can value the goodness
of a question form so that we could suggest right question form parts. In true earth, end-users are unready to make
ready clear and detailed take-back. The click-through on the make shorter view table is an if true, then some other is
necessarily true take-back to say to our system which cluster (or a division) of facts instances is desired by the User. The
clicked a division of is detailed by Duf. Note that Duf is only and a division of all User desired knowledge for
computers instances in the knowledge-base. But it can help our system produce recommended form parts that help
Users discover more desired facts examples. In some recommendation systems and look for engines, the end-users are
also let to make ready the not take-back. The not take-back is a group of the facts instances that are not desired by the
Users. In the question form outcomes, we take to be true most of the question facts instances are not desired by the
Users because if they are already desired, then the question forms complete persons living time is almost done. As an
outcome of that, the positive take-back is more giving knowledge than the not take-back in the question form complete
persons living time. Our made an offer design to be copied can be easily stretched for making into company the not
take-back. Made automatic position on scale of the results of a question is a pleasing to all point of view of the question
design to be copied in information acts to get back that we have grown to be dependent on. In contrast, knowledge-base
systems support only a Boolean question design to be copied. For example, a selection question on a SQL knowledgebase comes back all tuples that please the conditions in the question. As an outcome of that, the supporters two
scenarios are not with pleasing, smooth motion with a part for gripping by a SQL system. With nothing in answers:
When the question is too having selection, the answer may be with nothing in. In that example, it is desirable to have
the thing for which selection is made of requesting a position on scale list of approximately matching tuples without
having to specify the position on scale purpose, use that takes being near to the question. An Fbi person acting for or an
observer complex in knowledge for computers discovery will get such workings taking from lower to higher authority.
Many answers: When the question is not too having selection, too many tuples may be in the answer. In such an
Page 60

ISSN 2319 - 4847
example, it will be desirable to have the thing for which selection is made of ordering the matches automatically that
ranks more all over earth important answer tuples higher and coming back only the best matches. A person getting
goods from store taking grass for food a product price list will get such workings good-looking. Conceptually, the made
automatic position on scale of question results hard question is really that of taking a User question (say, a joining
selection question) and mapping it to a Top-K conditions in the User question. The key questions are: How to forming
of word from another such position on scale purposes, uses automatically? How well do position on scale purposes, uses
from send in name for? Are the position on scale techniques for putting one's hands on with nothing in answers and
many answers problems different? How to put to death such Top-K questions with small amount of money over greatly
sized knowledge-bases? We will start off by questioning ourselves how to make it possible for of relation knowledgebases to adjust position on scale group events from Ir for putting one's hands on the knowledge-base position on scale
hard question. When each property in the relation is an unlimited quality, we can person copying another Ir answer by
putting to use the TF-IDF idea that is based on the number of times of event of quality values in the knowledge-base.
However, unlike wording printed materials, knowledge-bases have within of numbers as well as unlimited information.
As an outcome of that, we need to stretch TF-IDF ideas of a quality common to a group to number fields (of
knowledge). We undergo growth IDF similarity, a knowledge-base position on scale purpose, use that gets stretched out
TF-IDF ideas of a quality common to a group to knowledge-bases having in it a heterogeneous mix of unlimited as well
as of numbers facts. While IDF similarity works well for some knowledge-base position on scale requests, sometimes
its good effect is quite limited. In certain instances the connection of knowledge for computers values for position on
scale may be needing payment to other factors in addition to their number of times. This has been noted in the lands
ruled over as well, where sometimes one has to go beyond TF-IDF weightings to forming of word from another
accurate position on scale group events. This begs the question: what else could be the base of general position on scale
in knowledge-bases? We make clear to that getting together the amount of work on the knowledge-base can be quite
useful for position on scale. In a way, this may be viewed as a poor mans good quality of connection take-back and
collaborative coming through slowly where a Users last good quality of on the point tuples is not recorded. Despite its
low in development nature, such amount of work information can help come to a decision about the number of times
with which knowledge-base properties and values are made, used statement, direction. When used in word used for
joining other words, statements with IDF, amount of work information makes stronger position on scale quality. We
undergo growth QF similarity, a position on scale purpose, use that leverages such amount of work information.
Much of the discussion in this paper gives one's mind to an idea on the with nothing in answers hard question. getting
answer to, way out of the many answers hard question puts forward added questions because a position on scale
purpose, use that only depends on the conditions in the User question is not (good) enough for this hard question. We
give (kind attention) our position on scale purposes, uses with added question independent parts that measure the
importance of tuples in a complete sense. At last, even if we get the position on scale purposes, uses right, for greatly
sized knowledge-bases; we have to make seem unimportant their force of meeting blow on question processing.
Although upside down lists are pleasing to all facts structures for good at producing an effect acts to get back in , they
are not (good) enough for our purposes as we look for imprecise matches getting mixed in trouble grouping and number
properties. We work-room adjustments of some nearby Algorithms for Top-K question processing, which leads us to
yet another something given of this paper; an index-based Top-K question processing Algorithm , ITA that great acts
our position on scale purposes, uses. We have made a system in which our position on scale Algorithms have been gave
effect to on an of relation DBMS. The system has two chief parts, a pre-processing part and a question processing part.
The preprocessing part is a position on scale group event extractor that leverages knowledge for computers and amount
of work qualities. The question processing part is a Top-K Algorithm that uses the position on scale purpose, use and
great acts the physical knowledge-base design. We have done User experiments 10 on our system to value its good
effect. However, despite our best attempts, our User experiments are preliminary. Unlike which is dependent on much
ready (to be used) User studies and benchmarks, no base structure is ready (to be used) today for valuing knowledgebase position on scale. The rest of this paper is put into order as takes as guide, example, and rule. In Section we have a
discussion related work. We make, be moving in two knowledge-base position on scale purposes, uses for the with
nothing in answers hard question, IDF similarity and QF similarity.
5. CONCLUSION
In this paper, we present a question recommendation framework supporting the effecting on one another discovery of
relation knowledge-bases and an instantiation of this framework based on user-based collaborative coming through
slowly. The testing put value makes clear by reasoning the possible & unused quality of the made an offer move near.
We should conditions making things hard that this is a first-cut answer to the very interesting hard question of made
for a person question statements of good words for. There are many open issues that need to be made house numbers.
For example, an interesting hard question is that of making out similar questions in terms of their structure and not the
tuples they put in good order again two questions might be semantically similar 1 but get back different results needing
payment to some coming through slowly conditions. Such questions need to be thought out as in the recommendation
Page 61

ISSN 2319 - 4847
process. We are currently working on getting stretched out our framework to cover such question similarities. Another
interesting direction is to put to use item-based collaborative coming through slowly instead of the user-based move
near of the current framework. We also make up one's mind to have a look for other moves near for instantiating the
made an offer with the idea framework.
REFERENCES
[1]
[2]
[3]
[4]
[5]
Cold Fusion. http://www.adobe.com/products/coldfusion/.

DBPedia. http://DBPedia.org.
Easy Query. http://devtools.korzh.com/eq/dotnet/.
Freebase. http://www.freebase.com.
C. C. Aggarwal, J. Han, J. Wang, and P. S. Yu. A framework for clustering evolving data streams. In Proceedings
of VLDB, pages 8192, Berlin, Germany, September 2003.
[6] R. Agrawal, S. Gollapudi, A. Halverson, and S. Ieong. Diversifying search results. In Proceedings of WSDM,
pages 514, Barcelona, Spain, February 2009.
Page 62

Dynamic Generation For System Over Database

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dynamic Generation For System Over Database

Uploaded by

Copyright:

Available Formats

International Journal of Application or Innovation in Engineering & Management (IJAIEM)

Web Site: www.ijaiem.org Email: editor@ijaiem.org

ISSN 2319 - 4847