International Journal of Engineering Trends and Technology (IJETT) Volume 4 Issue 9- Sep 2013
ISSN: 2231-5381 http://www.ijettjournal.org Page 4227
Personalized Query Results using User
Search Logs
Santhi Kolli
1
V.Ramachandran
2
Dr.Rupa
3
1
PG Student (M.Tech CSE), Vasireddy Venkatadri institute of Technology, Guntur, AP, India
2
Research Scholor, Acharya Nagarjuna University, Nambur, Guntur, AP, India
3
Professor & H.O.D, Vasireddy Venkatadri institute of Technology, Guntur, AP, India
ABSTRACT:
Most users want their search engine to incorporate three key features in query results. Relevant results (results they
are actually interested in), Uncluttered (easy to read interface), Helpful options to broaden or tighten a search for
accuracy. This paper addresses the third aspect with new improvement measures for an enhanced experience to the
end user. A trivial query like purchasing a laptop has to be broken down into a number of co-dependent steps over a
period of time based on prior search patterns of the same user .For instance, a user may first search for company and
later the features and price. After deciding the company and price the user may then search for the accessories like
mouse and modem etc. Each step requires more queries and each query completes with more clicks. Current search
engines cannot support this kind of hierarchical queries. We propose to implement Random walk propagation
methods that can construct user profiles based on the credentials obtained from their prior search history
repositories. Combined with click points driven click graphs of user search behavior the IR system can support
complex queries for future requests at reduced navigations. Random walk propagation over the query fusion graph
methods support complex search quests in IR systems at reduced times. For developing an interactive IR system we
also propose to use these search quests as auto complete features in similar query propagations. Biasing the ranking
of search results can also be provided using ranking algorithms (top-k algorithms).Supporting these methods yields
dynamic and improved performance in IR systems, by providing enriched user querying experience. A practical
implementation of the proposed system validates our claim.
I. INTRODUCTION
Web search engines attempt to satisfy users
information needs by ranking web pages with respect
to queries. AS the size and richness of information on
the Web grows, hence the variety and the complexity
of tasks that users try to accomplish online. Only
20% of the queries are the navigational on the various
analyses of query logs and the rest are transactional
or advisory in the nature. A series of interactions
between user and search engine can be necessary to
satisfy a single information need.
International Journal of Engineering Trends and Technology (IJETT) Volume 4 Issue 9- Sep 2013
ISSN: 2231-5381 http://www.ijettjournal.org Page 4228
The Internet searches are a top Internet
activity, second only to email in the report of the web
search. The information retrieval is the important task
doing by the users The search engine is still through
the keyword queries to access information online.
Over a period of time the complex tasks such as the
purchasing a laptop is divided into many steps.
Although some websites such as the redbus, ebay are
helpful to provide from the single database we can
attain the steps. So the user want to search the
company,features and price. After the selection of the
company user searches for accessories like mouse
and modem etc.Each step requires more than one
query and each query requires more clicks on the
relevant pages.
During the online searching group the
related queries need to be grouped together.All the
search engines are using using the Search History
where users can allow tracking their online searches
by recording their queries and clicks. The figure1 can
illustrate the portion of users history as it shown by
the google. In addition to viewing their search
history, users can manipulate it by manually editing
and organizing related queries and clicks into groups,
or by sharing them with their friends.. Identifying
groups of related queries has applications beyond
helping the users to make sense and keep track of
queries and clicks in their search history.
Fig 1: Example of search history feature in Google
Query grouping takes users session. The search
engines are considered to be good if they identify
query groups and the clicks in the query groups. The
components of search engines such as the result
ranking, sessionization, query suggestions,
collaborative search and query alterations. Consider a
search engine knows that a current query Mac
belongs to the {Apple, Mac} group query. The
rank of the page can be improved which provides
information about how to get a Mac instead of article
on Apple. Query grouping can also helps other
users by promoting task-level collaborative search.
Explicit collaborative search can also be performed
by allowing users in a trusted community to combine,
find and share relevant query groups to perform
larger, long-term tasks on the Web.
The query group has been arrange in the automated
and dynamic fashion in our report of organizing the
users search history. Each query group is a
collection of queries by the same user that are
relevant to each other around a common
informational need. When the user issues a new
International Journal of Engineering Trends and Technology (IJETT) Volume 4 Issue 9- Sep 2013
ISSN: 2231-5381 http://www.ijettjournal.org Page 4229
query then
query
groups can
automatic
ally
updated. To illustrate our intension let see the below
tables1, the related queries Apple and Apple
IPhone are separated by many unrelated queries.
Time Query Time Query
11:51:48
Aeroplane
12:59:12
APPSC
Material
12:52:24
Age Calculator
13:03:34
APPSC
Notification
01:59:28
APJ
Abdulkalam
Quotes
16:34:09
Axis Bank
11:25:04
APJ
17:52:49
Acharya
Abdulkalam
Speech
Nagarjuna
University
11:57:23
Age
19:22:13
Acharya
Nagarjuna
University
Results
12:00:02
Apple
19:25:49
Apple Fruit
11:40:27
Apple India
19:50:12
Apple Iphone
12:32:42
APPSC
20:11:56
Apple Free
Apps
12:22:22
APPSC Results
20:44:01
Apple Mac
(a) Users Search History
Group 1
APJ Abdulkalam
APJ Abdulkalam Speech
APJ Abdulkalam Quotes
Group 2
Apple
Apple Fruit
Apple Free Apps
Apple I phone
Apple Mac
Group 3
Acharya Nagarjuna University
Acharya Nagarjuna University
Results
International Journal of Engineering Trends and Technology (IJETT) Volume 4 Issue 9- Sep 2013
ISSN: 2231-5381 http://www.ijettjournal.org Page 4230
(b) Query Groups
Table 1: Search history of a real user over the
period of one day together with the query groups
But, the similar quires may not be textually similar.
Consider in the example table1 (b) the related queries
trip advisor Barbados and Caribbean cruise in
Group 2 have no words in common. Finally, as users
may also manually alter their respective query
groups, the users can edit or manual effort of the user
can automate by query grouping.
To achieve more effective and robust query
grouping, we do not rely solely on textual or
temporal properties of queries. We leverage search
behavioral data as captured within a commercial
search engines log. In general, we develop an online
query grouping
method over the query fusion graph that combines a
probabilistic query reformulation graph. Related to
our problem, the problems of session identication,
and query clustering, that have also used similar
graphs in the past. In our scheme we make the
following contributions:
Our design is to clinch good performance
while avoiding disruption of existing user-
dene query groups.
We analysis two potential ways of using
clicks in order to enhance this process:
1. Query fusion graph is referred as
query click graph into a single
graph by fusing the query
reformulation graph.
2. The query set when computing
relevance to also include other
queries with similar clicked URLs
on the expanding.
Our proposed search log-based method
shows the effectiveness and the robustness
by comprehensive experimental evaluation.
II. PRELIMINARIES
The aim is to organize a users search logs into query
groups,which contains one or more related queries
The query group consists of atomic information from
the queries. For broader informational queries the
query group consists of a sub queries and clicks we
may see at the table1 (b) group 5.
Query Group: A query group is an ordered list of
queries, q
i
, together with the corresponding set of
Group 4
Age
Age Calculator
Group 5
APPSC
APPSC Results
APPSC Notification
International Journal of Engineering Trends and Technology (IJETT) Volume 4 Issue 9- Sep 2013
ISSN: 2231-5381 http://www.ijettjournal.org Page 4231
clicked URLs, clk
i
of q
i
. A query group is denoted as s
= h{q
1
, clk
1
}, . . . , {q
i
, clk
i
}.
The query group may be either an existing query
groups or a new query group if the query group does
not exists.
Query group is used to measure the relevance
between two queries.
Dynamic Query Grouping: The identication of
query groups is the important task. For this consider
the query group as a singleton query group, and
singleton query can be combined iteratively.. First,
changing a users existing query groups effects the
users history. Second, it involves a high
computational cost.
The grouping in done by using online clustering
algorithms. First place the current query and clicks
into a singleton query group sin
c
={q
c
, clk}. Compare
it with each existing query group s
i
within a users
history (i.e.,s
i
S). Identify if there are existing
query groups which are relevant to s
c
.
Query Relevance:
To verify whether the query group contains related
queries and clicks. The measure sim between the
current query singleton group s
c
and an existing query
group s
i
S.. we assume that users give similar
queries and clicks within a short period of time. can
The following time-based relevance metric sim
time
that can be used in place of sim in figure2.
SelectBestQueryGroup
Input:
1) The current singleton query group s
c
containing the
Current query q
c
and set of clicks dk
c
2) A set of existing query group S ={s
1
,.,s
m
}
3) A similarity threshold T
sim
, 0 <T
sim
<1
Output: The query group s that best matches s
c
, or a
new
One if necessary
(0) S=0
(1) T
max
=T
sim
(2) for i =1 to m
(3) if sim(s
c
,s
i
) >T
max
(4) s =s
i
(5)
T
max
=sim(s
c
,s
i
)
(6) if s =0
(7) S =S U s
c
(8) s =s
c
(9) return s
Fig.2: Algorithm for selecting the query group that is
the most similar to the given query and clicked URLs.
The multitasking is not considered in this scenario..
The relevance between query groups around textually
similar queries such as Apple and AppleIphone.
It may not give relevancy measure between the query
groups around queries such as Apple and Apple
Iphone. The user search logs are used in order to
determine the relevance between query groups. To
measure the relevance between query groups query
logs and the click logs are used.
III. QUERY RELEVANCE USING
SEARCH LOGS
Query Relevance means developing the system
based on Web search. The works defines the main
properties of relevant queries.
1. Frequent queries which appear together as
reformulations
International Journal of Engineering Trends and Technology (IJETT) Volume 4 Issue 9- Sep 2013
ISSN: 2231-5381 http://www.ijettjournal.org Page 4232
2. queries which forces the users to click on
similar pages.
These graphs are used to compute query relevance
and the way of incorporating the clicks of a users
query to enhance our relevance metric.
3.1 Search Behavior Graphs
The search history of a search engine can categorized
into three types of graphs.
1. query reformulation Graph - represents the
relationship between a pair of queries
that are likely reformulations of each
other
2. query click graph - forces users to click on
same URLs.
3. query fusion graph - merges the
information in the previous two graphs.
1. Query Reformulation Graph
If two queries that are issued at the same
time by many users they are likely to be similar
to each other. The relevance of two queries can
be measured by the time metric (S
t
) means
interval between the timestamps of the queries.
Our main aim is to dene the statistical
frequency of two queries which appear next to
each other over all of the users of the system.
The query reformulation graph is
QRG =(V
q
, E
q
)
E
q
=set of edges, constructed as per the
query sets (q
i
,q
j
) , where we can issue the q
i,
before the q
j
with in the users day activity.,
we remove infrequent query pairs and
include only the query pairs whose counts is
more than threshold value ( T
v
).
The edge weight, W
r
(q
i
,q
j
), is the
normalized count of the query transitions:
W
r
(q
i
,q
j
) = cnt
(q
i
,q
j
)
(q
i
,q
k
)E
q
cnt
(q
i
,q
k
)
2. Query Click Graph
Take queries differently to capture
relevant queries from the search historyt are
likely to force users to click frequently on the
same set of URLs. Consider the queries Mac
and Apple do not share any text and close to
each other..They are relevant to each other
because they give the same results.
The bipartite click-through graph, CG =
(Vm Vn, Eg). The infrequent pairs are removed
using a threshold T
g
.
w
g
(q
i
,q
j
) =
uk
min (cnt
g
(q
i
,u
k
),cnt
g
(q
j
,u
k
)
uk
cnt
g
(q
i
,q
k
)
3. Query Fusion Graph
The query reformulation graph (QRG) and
the Query fusion graph (QFG) shows the
properties of relevant queries.The query
reformulation information is combined with
QRG and the query click information is
combined with QCG into a single graph QFG=
(V
f
, E
fg
). The weight of edge (q
i
,q
j
) in QFG,
International Journal of Engineering Trends and Technology (IJETT) Volume 4 Issue 9- Sep 2013
ISSN: 2231-5381 http://www.ijettjournal.org Page 4233
Wf(q
i
,q
j
), is the sum of the weights, Wr(q
i
,q
j
)
in EQR and Wg(q
i
,q
j
) in EQC.
W
f
(q
i
,q
j
) = X W
r
(q
i
,q
j
) +(1 ) X
Wg(q
i
,q
j
)
3.2 Computing Query Relevance
For a user query q, the relevance vector is
calculated using QFG, where each entry represents
the relevance value of each query qj VQ to q.
Because this is not sufficient to use the pair wise
relevance values in QFG a more sophisticated called
Markov chain for q (MKq) is used.This is called the
fusion relevance vector of q rel
fq
for stationary
distribution.
Many users issuing queries and clicks in
real-time and the huge size of QFG. Relavance(q)
Input:
1) Query FusionGgraph, QFG
2) Jump Vector, j
3) Damping Factor, d
4) Number of Random Walks, noRW
s
5) Size of Neighborhood, mxHops
6) the given query, q
Output:Fusion Relevance Vector for q, rel
Fq
(0) Initialize rel
Fq
=0
(1) noWaks =0; noVisits =0
(2) while noWaks <noRWs
(3) noHops =0; v=q
(4) While v NULL noHops <mxHops
(5) noHops++
(6) rel
Fq
(v)++; noVists++
(7) v = SelectNextNodeToVisit(v)
(8) noWalks++
(9) For each v, normalize rel
Fq
(v) =
rel
Fq
(v)/noVisits
Fig. 3. Algorithm for the query relevance
The algorithm computes the fusion
relevance vector of a given query q rel
fq
. It requires
the following inputs in addition to QFG. Two other
inputs for the accuracy and the time budget are
1.Total number of random walks (noRWs).
2.The size of neighborhood (maxHops).
Tthe length of each random walk is maxHops, if the
queries from q to q
l
are not alike.
The number of random walk or the lengths of random
walks are limited by decreasing both parameters for a
faster computation of rel
fq
. . The selection of the
next node is based on the outgoing edges of the
current node v in QFG. The damping factor d is
performed by the SelectNextNodeToVisit process in
Step (7) of the algorithm, it can be shown in the
figure 4.
SelectNextNodeToVisit(v)
Input:
1.Query fusion graph, QFG
2. J ump Vector, j
3.Damping Factor, d
4. Current Node, v
Output: Next Node to visit, q
i
(0) if rdm() <d
(1) V ={q
i
| (v,q
i
) QF }
(2) Pick a node q
i
V with probability
w
f
(v,q
i
)
(3) else
(4) V ={q
i
| j(q
i
) >0}
(5) Pick a node q
i
V with probability
g(q
i
)
(6) Return q
i
Fig. 4. Algorithm to select the next node to visit.
International Journal of Engineering Trends and Technology (IJETT) Volume 4 Issue 9- Sep 2013
ISSN: 2231-5381 http://www.ijettjournal.org Page 4234
IV. QUERY GROUPING
The relevance of other queries is captured in
an image. The context vector is noted which
aggregates the images of queries to form an overall
representation.. The query reformulation graph, query
images, and context vectors are used in which lend
Markov chain process to determine relevance
between queries and query groups.
Context Vector: A context vector is used to
calculate the similarity between the query group and
the users query group. The query group s of the
context vector is denoted as the cxt
s.
The context
vector of a query group s by weighting the queries
and the clicks in s by recency, as follows:
cxt
s
=
recency
recency
)
k
j
rel(qs1,clk1)
Query Image: consider two relevant queries,
Apple and Apple Mac in table 1. The relevance
value in the fusion relevance vectors rel
Apple
(Apple
Mac)
.
Online Query Grouping: To avoid performing the
random walk computation of fusion relevance vector
for each query estimate that to cache the fusion
relevance vectors of 100 million queries..
V. EXISTING SYSTEM
A user queries a search engine has been
existed. The search engine displays results based
page ranking algorithms.. The primary means of
accessing information online is still through keyword
queries to a search engine. A complex task such as
Buying a laptop has to be broken down into a
number of co-dependent steps over a period of time.
For example, a user may first search on the company,
features of laptops, prices etc. After deciding the
company and prices the user may search for the most
accessories like mouse and modem.Each step
requires one or more queries, and each query results
in one or more clicks on relevant pages. Keyword
based search engines cannot address this kind of
complicated tasks. So a better system is required that
can enable the user to pursue complex search quests
online. Search Engine tries to construct user profile
based on his ipaddress/login credentials from its user
search history repositories. If the user already exists,
the search engine checks from its user search history
repositories up to a certain threshold whether the user
already queried the same query previously. If the user
did, then search engine further retrieves click points
from user search history repositories and
reformulates query results by generating click graphs.
Click graphs contain useful information on user
behavior when searching online. This step is called
query fusion graph. Uses random walk propagation
over the query fusion graph instead of time-based and
keyword similarity based approaches. This entire
process is called Personalizing the user search
histories into query groups.
VI. PROPOSED SYSTEM
Most users want their search engine to
incorporate three key features in query results. A
International Journal of Engineering Trends and Technology (IJETT) Volume 4 Issue 9- Sep 2013
ISSN: 2231-5381 http://www.ijettjournal.org Page 4235
query such as buying a laptop has to be divided into
different steps.iew. For example, a user may first
search on the company, features of laptops, prices
etc. After deciding the company and prices the user
may search for the most accessories like mouse and
modem. The click points of click graphs of user
search behavior the IR system can support complex
queries for future requests at reduced navigations.
Random walk propagation over the query fusion
graph methods support complex search quests in IR
systems at reduced times. For developing an
interactive IR system we also propose to use these
search quests as auto complete features in similar
query propagations. Biasing the ranking of search
results can also be provided using ranking algorithms
(top-k algorithms).These methods gives improved
performance in IR systemsA practical
implementation of the proposed system validates our
claim. During the online complex quest to identify
and to group the related queries together we have a
standout step towards the enabling services and
feature that are capable. Currently we are using the
Search History in major search engines where users
can allow tracking their online searches by recording
the queries and clicks.
VII. CONCLUSION
Click graphs have information on user behavior
while searching online. This step is called query
fusion graph. Uses random walk propagation over the
query fusion graph instead of time-based and
keyword similarity based approaches. This entire
process is called personalizing user search histories
into query groups. For making the IR Systems
effective and dynamic we propose to use these search
quests as auto complete features in similar query
propagations. Biasing the ranking of search results
can also be provided using any ranking
algorithms(top-k algorithms). Supporting these
methods yields dynamic performance in IR systems,
by providing enriched user querying experience
VIII. REFERENCES
[1] Heasoo Hwang, Hady W. Lauw, Lise Getoor and
Alexandros Ntoulas, Organizing User Search
Histories, Volume: 24 , Issue: 5 in IEEE 2012
Transactions on Knowledge and Data Engineering
[2] J. Teevan, E. Adar, R. Jones, and M. A. S. Potts,
Information reretrieval: repeat queries in yahoos
logs, in SIGIR. New York, NY, USA: ACM, 2007,
pp. 151158.
[3] A. Broder, A taxonomy of web search, SIGIR
Forum, vol. 36, no. 2, pp. 310, 2002.
[4] R. Jones and K. L. Klinkner, Beyond the session
timeout: Automatic hierarchical segmentation of
search topics in query logs, in CIKM, 2008.
[5] A. Spink, M. Park, B. J. Jansen, and J. Pedersen,
Multitasking during Web search sessions,
Information Processing and Management, vol. 42,
no. 1, pp. 264275, 2006.
[6] A. Spink, B. J. Jansen, and H. C. Ozmultu. Use of
query reformulation and relevance feedback by
Excite users. Internet Research: Electronic
International Journal of Engineering Trends and Technology (IJETT) Volume 4 Issue 9- Sep 2013
ISSN: 2231-5381 http://www.ijettjournal.org Page 4236
Networking Applications and Policy, 10(4):317328,
2000.
Authors Profile
Santhi Kolli
She completed her Masters Computer Science from
Periyar University.Right now she is the student of
Vasireddy Venkatadri institute of Technology
Pursuing M.Tech in Computer Science and
Engineering.
Ramachandran, is a research scholar at Acharya
NagarjunaUniversity, Nambur. He got his B.TECH
Computer Science &Systems engineering
Degreefrom Andhra University andM.TECH in
Computer ScienceEngineering from JNTU,
Kakinada. He is verymuch interested in image
processing, medical retrieval , human vision &
pattern recognition.He did several projects in image
processing.
Dr Rupa, working as theof the Department of
Computer Science and Engineering in
VVIT.Nambur.She has 12 yearsof teaching
experience. Dr Ch.Rupa, obtained her
B.Tech(CSIT) Degree in Computer Science &
Information Technology fromJ NTU, Hyderabad
in 2002 and M.Tech (IT) in Information
Technology from Andhra University in 2005.
She
was awarded Ph. D (CSE) in Computer Science
Engineering by Andhra University.
