Sourav S. Bhowmick

Research Issues in Web Data Mining

In this paper, we discuss mining with respect to web data referred here as web data mining. In particular, our focus is on web data mining research in context of our web warehousing project called WHOWEDA (Warehouse of Web Data). We have categorized web data mining into threes areas; web content mining, web structure mining and web usage mining. We have highlighted and discussed various research issues involved in each of these web data mining category. We believe that web data mining will be the topic of exploratory research in near future.

Mobile data and transaction management

Abstract Mobile computing paradigm has emerged due to advances in wireless or cellular networking technology. This rapidly expanding technology poses many challenging research problems in the area of mobile database systems. The mobile users can access information independent of their physical location through wireless connections. However, accessing and manipulating information without restricting users to specific locations complicates data processing activities. There are computing constraints that make mobile database processing different from the wired distributed database computing. In this paper, we survey the fundamental research challenges particular to mobile database computing, review some of the proposed solutions and identify some of the upcoming research challenges. We discuss interesting research areas, which include mobile location data management, transaction processing and broadcast, cache management and replication and query processing. We highlight new upcoming research directions in mobile digital library, mobile data warehousing, mobile workflow and mobile web and e-commerce.

Time-dependent semantic similarity measure of queries using historical click-through data

It has become a promising direction to measure similarity of Web search queries by mining the increasing amount of click-through data logged by Web search engines, which record the interactions between users and the search engines. Most existing approaches employ the click-through data for similarity measure of queries with little consideration of the temporal factor, while the click-through data is often dynamic and contains rich temporal information. In this paper we present a new framework of time-dependent query semantic similarity model on exploiting the temporal characteristics of historical click-through data. The intuition is that more accurate semantic similarity values between queries can be obtained by taking into account the timestamps of the log data. With a set of user-defined calendar schema and calendar patterns, our time-dependent query similarity model is constructed using the marginalized kernel technique, which can exploit both explicit similarity and implicit semantics from the click-through data effectively. Experimental results on a large set of click-through data acquired from a commercial search engine show that our time-dependent query similarity model is more accurate than the existing approaches. Moreover, we observe that our time-dependent query similarity model can, to some extent, reflect real-world semantics such as real-world events that are happening over time.

Event detection from evolution of click-through data

Previous efforts on event detection from the web have focused primarily on web content and structure data ignoring the rich collection of web log data. In this paper, we propose the first approach to detect events from the click-through data, which is the log data of web search engines. The intuition behind event detection from click-through data is that such data is often event-driven and each event can be represented as a set ofquery-page pairs that are not only semantically similar but also have similar evolution pattern over time. Given the click-through data, in our proposed approach, we first segment it into a sequence of bipartite graphs based on theuser-defined time granularity. Next, the sequence of bipartite graphs is represented as a vector-based graph, which records the semantic and evolutionary relationships between queries and pages. After that, the vector-based graph is transformed into its dual graph, where each node is a query-page pair that will be used to represent real world events. Then, the problem of event detection is equivalent to the problem of clustering the dual graph of the vector-based graph. The clustering process is based on a two-phase graph cut algorithm. In the first phase, query-page pairs are clustered based on thesemantic-based similarity such that each cluster in the result corresponds to a specific topic. In the second phase, query-page pairs related to the same topic are further clustered based on the evolution pattern-based similarity such that each cluster is expected to represent a specific event under the specific topic. Experiments with real click-through data collected from a commercial web search engine show that the proposed approach produces high quality results.

A Model for XML Schema Integration

We define an object-oriented data model called XSDM (XML Schema Data Model) and present a graphical representation of XML Schema integration. The three layers included are, namely, pre-integration, comparison and integration. During pre-integration, the schema present in XML Schema notation is read and is converted into the XSDM notation. During the comparison phase of integration, correspondences as well as conflicts between elements are identified. During the integration phase, conflict resolution, restructuring and merging of the initial schemas take place to obtain the global schema.

An XML Schema integration and query mechanism system

The availability of large amounts of heterogeneous distributed web data necessitates the integration of XML data from multiple XML sources for many reasons. For example, currently, there are many e-commerce companies, which offer similar products but use different XML Schemas with possibly different ontologies. When any two such companies merge, or make an effort to service customers in cooperation, there is a need for an integrated schema and query mechanism for the interoperability of applications. In applications like comparison-shopping, there is a need for an illusionary centralized homogeneous information system. In this paper, we propose XML Schema integration and querying methodology. We define an object-oriented data model called XSDM (XML Schema Data Model) and present a graphical representation of XML Schema for the purpose of schema integration. We use a three-layered architecture for XML Schema integration. The three layers included are namely pre-integration, comparison, and integration. The three layers can conceptually be regarded as three phases of the integration process. During pre-integration, the schemas present in XML Schema notation are read and converted into the XSDM notation. During the comparison phase of integration, correspondences as well as conflicts between elements are identified. During the integration phase, conflict resolution, restructuring and merging of the initial schemas takes place to obtain the global schema. We define integration policies for integrating element definitions as well as their datatypes and attributes. An integrated global schema forms the basis for querying a set of local XML documents. We discuss various strategies for rewriting the global query over the global schema into the sub-queries over local schemas. Their respective local schemas validate the sub-queries over the local XML documents. This requires the identification and use of mapping rules and relationships between the local schemas.

Image tag clarity: in search of visual-representative tags for social images

Tags associated with images in various social media sharing web sites are valuable information source for superior image retrieval experiences. Due to the nature of tagging, many tags associated with images are not visually descriptive. In this paper, we propose Normalized Image Tag Clarity (NITC) to evaluate the effectiveness of a tag in describing the visual content of its annotated images. It is measured by computing the zero-mean normalized distance between the tag language model estimated from the images annotated by the tag and the collection language model. The visual-representative tags that are commonly used to annotate visually similar images are given high tag clarity scores. Evaluated on a large real-world dataset containing more than 269K images and their associated tags, we show that NITC score can effectively identify the visual-representative tags from all tags contributed by users. We also demonstrate through experiments that most popular tags are indeed visually representative.

Web warehousing: an algebra for web information

While conventional keyword indexes maintained by web search engines such as Yahoo, Lycos, and World Wide Web Worm work well for most simple keyword searches, they are inadequate when more complex and structured queries involving the underlying hypertext structure of the World Wide Web are desired. Building from a database perspective, existing work to support such queries focuses on constructing SQL-like query languages for the WWW that assumes a relational abstraction of the WWW. Nonetheless, the WWW is a directed graph and imposing a relational abstraction filters out its inherent topological structure. We propose a data model for the WWW that retains its topological structure and construct a web algebra to manipulate objects in this model. The web algebra establishes a formal foundation from which different web query languages can be designed.

DTD-Diff: A change detection algorithm for DTDs

The DTD of a set of XML documents may change due to many reasons such as changes to the real-world events, changes to the users requirements, and mistakes in the initial design. In this paper, we present a novel algorithm called DTD-Diff to detect the changes to DTDs that defines the structure of a set of XML documents. Such change detection tool can be useful in several ways such as maintenance of XML documents, incremental maintenance of relational schema for storing XML data, and XML schema integration. We compare DTD-Diff with existing XML change detection approaches and show that converting DTD to XML schema (XSD) (which is in XML document format) and detecting the changes using existing XML change detection algorithms is not a feasible option. Our experimental results show that DTD-Diff is 5-325 times faster than X-Diff when it detects the changes to the XSD files. Compared to XyDiff, DTD-Diff is up to 38 times faster. We also study the result quality of detected deltas.

Cell-delivery therapeutics for liver regeneration☆

For acute, chronic, or hereditary diseases of the liver, cell transplantation therapies can stimulate liver regeneration or serve as a bridge until liver transplantation can be performed. Recently, fetal hepatocytes, stem cells, liver progenitor cells, or other primitive and proliferative cell types have been employed for cell transplantation therapies, in an effort to improve the survival, proliferation, and engraftment of the transplanted cells. Reviewing earlier studies, which achieved success by transplanting mature hepatocytes, we propose that there is a switch-like regulation of liver regeneration that changes state according to a stimulus threshold of extracellular influences such as cytokines, matrices and neighboring cells. Important determinants of a successful clinical outcome include sufficient quantities and functional levels of the transplanted cells (even for short periods to alter the environment), rather than just engraftment levels or survival durations of the exogenously transplanted cells. The relative importance of these determining factors will impact future choices of cell sources, delivery vehicles, and sites of cell transplantation to stimulate liver regeneration for patients with severe liver diseases.