|
Sept. 29th, 2006
13:00-14:00
|
Jing Zhao
|
Title:
Improving Meta-search Quality Using Distributed Latent Semantic Indexing Abstract:
The trade-off between improving search quality and minimizing search engine summarization cost has been one of the most important issues in a meta-search system. We proposed a search engine descriptor based on distributed latent semantic indexing. The descriptor represented the semantics of a search engine's document collection which are characterized by term co-occurrences across documents. Furthermore, we divided a document collection into clusters by indexed terms and did latent semantic indexing within clusters so as to improve the effectiveness and efficiency of search engine summarization. We also proposed result merging algorithm based on latent semantic indexing. Our experimental results over the TREC collection showed that our server selection method significantly improved search precision with a low storage and update cost. In addition, our result merging algorithm was verified as effective especially when a small number of top-ranked results were examined. |
|
|
Oct 20th, 2006,
13:00-14:00
Room: 3530
|
Su WeiFeng
|
Title: Automatic Hierarchical Classification of Structured DeepWeb Databases Abstract: We present a method that automatically classifies structured deepWeb databases according to a pre- defined topic hierarchy. We assume that there are some manually classified databases, i.e., training databases, in every node of the topic hierarchy. Each training database is probed using queries constructed from the node titles of the topic hierarchy and the query result counts reported by the database are used to represent the content of the database. Hence, when adding a new database it can be probed by the same set of queries and classified to a node whose training databases are most similar to the new one. Specifically, a support vector machine classifier is trained on each internal node of the topic hierarchy with these training databases and the new database can be classified into the hierarchy top-down level by level. A feature extension method is proposed to create discriminate features. Experiments run on real structured Web databases collected from the Internet show that this classification method is quite accurate. |
|
|
Nov 24th, 2006
13:00-14:00
Room: 3464
|
Gabriel Ghinita
Ph.D.
Candidate
Department
of Computer Science, NUS
|
Title: Privacy-Preserving Spatial Queries in Location Based Services Abstract:
The emerging trend of mobile
devices with embedded positioning capabilities (e.g., GPS) facilitates the
widespread use of Location Based Services. For such applications to succeed,
query privacy and confidentiality are of paramount importance. Conventional
privacy-preserving techniques include encryption, which safeguards communication
channels, and the use of pseudonyms, which hide user identities.
Nevertheless, the contents of spatial queries may disclose the physical
location and identity of users, compromising their privacy. We present
a framework for preserving the privacy of users who issue spatial queries to
Location Based Services. We propose transformations based on the
well-established k-anonymity paradigm to compute exact answers for Range and
Nearest Neighbor queries, without revealing the query source identity. Our proposed
techniques provide guarantees on user privacy, and can be employed both in
the centralized setting, as well as in decentralized/P2P systems. Extensive
experimental studies show that our methods are applicable to real-life
scenarios with numerous mobile users.
|
|
|
Dec 18th, 2006
14:00-15:00
Room: 3530
|
Huiming Qu
Ph.D candidate
Department of Computer Science
University of Pittsburgh
and
Jimeng Sun
Ph.D candidate
Computer Science Department
Carnegie Mellon University
|
Title: Preference-Aware Query and Update Scheduling for Web-Databases By Ms. Huiming Qu Abstract:
Web-database systems are
nowadays an integral part of everybody's life. In general, users expect short
response times and low staleness. However, it may be extremely hard to apply
all updates on time, i.e., keep zero staleness, and also get fast response
times, especially in periods of bursty traffic. In this work, we present the
concept of Quality Contracts (QCs) which combines the two incomparable
performance metrics: response time or Quality of Service (QoS), and staleness
or Quality of Data (QoD). QCs allows individual users to express their
preferences for the expected QoS and QoD of their queries by assigning "
profit" values. To maximize the total profit from submitted QCs, we
propose an adaptive algorithm, called QUTS. QUTS addresses the problem of
prioritizing the scheduling of updates over queries using a meta scheduling
scheme that dynamically allocates CPU resources to updates and queries
according to user preferences. We present the results of an extensive
experimental study using real data (taken from a stock information web site),
where we show that QUTS performs better than baseline algorithms under the
entire spectrum of QCs; QUTS also adapts fast to changing workloads.
Title: Window-based Tensor Analysis on High-dimensional and Multi-aspect Streams By Mr. Jimeng Sun Abstract:
Data stream values are often
associated with multiple aspects. For example, each value from environmental
sensors may have an
associated type (e.g.,
temperature, humidity, etc) as well as location. Aside from timestamp, type
and location are the two additional
aspects. How to model
such streams? How to simultaneously find patterns within and across the
multiple aspects? How to do it
incrementally in a
streaming fashion? In this paper, all these problems are addressed through a
general data model, tensor streams, and
an effective
algorithmic framework, window-based tensor analysis (WTA). Two variations of
WTA, independentwindow tensor
analysis (IW) and
moving-window tensor analysis (MW), are presented and evaluated extensively
on real datasets. Finally, we illustrate
one important
application, Multi-Aspect Correlation Analysis (MACA), which uses WTA and we
demonstrate its effectiveness on an
environmental
monitoring application.
Remark: similar talk will be given at ICDM06.
|
|