An Evidential Approach to Query Interface Matching on the Deep Web

Matching query interfaces is a crucial step in data integration across multiple Web databases. The problem is closely related to schema matching that typically exploits different features of schemas. Relying on a particular feature of schemas is not suffcient. We propose an evidential approach to combining multiple matchers using Dempster-Shafer theory of evidence. First, our approach views the match results of an individual matcher as a source of evidence that provides a level of confidence on the validity of each candidate attribute correspondence. Second, it combines multiple sources of evidence to get a combined mass function that represents the overall level of confidence, taking into account the match results of different matchers. Our combination mechanism does not require use of weighing parameters, hence no setting and tuning of them is needed. Third, it selects the top k attribute correspondences of each source attribute from the target schema based on the combined mass function. Finally it uses some heuristics to resolve any conflicts between the attribute correspondences of different source attributes. Our experimental results show that our approach is highly accurate and effective.

Visually extracting data records from the deep web

Web sites that rely on databases for their content are now ubiquitous. Query result pages are dynamically generated from these databases in response to user-submitted queries. Automatically extracting structured data from query result pages is a challenging problem, as the structure of the data is not explicitly represented. While humans have shown good intuition in visually understanding data records on a query result page as displayed by a web browser, no existing approach to data record extraction has made full use of this intuition. We propose a novel approach, in which we make use of the common sources of evidence that humans use to understand data records on a displayed query result page. These include structural regularity, and visual and content similarity between data records displayed on a query result page. Based on these observations we propose new techniques that can identify each data record individually, while ignoring noise items, such as navigation bars and adverts. We have implemented these techniques in a software prototype, rExtractor, and tested it using two datasets. Our experimental results show that our approach achieves significantly higher accuracy than previous approaches. Furthermore, it establishes the case for use of vision-based algorithms in the context of data extraction from web sites.

Isotopic variation complicates analysis of trophic relations within the fish community of Plußsee: a small, deep, stratifying lake

Analysis of carbon and nitrogen stable isotopes has allowed freshwater ecologists to examine lake food webs in increasing detail. Many such studies have highlighted the existence of separate within-lake pelagic and benthic-littoral food webs but are typically conducted on large (> 10 km2) lakes, whereas the majority of lakes are actually relatively small. We used stable isotope analysis (δ13C & δ15N) to examine trophic interactions between fish and their prey in Plu�see, as an example of a small, stratifying lake, and to determine whether separate pelagic/benthic-littoral food webs could be distinguished in such systems. Our results indicate that the Plu�see food web was complicated, and due to extensive intra-annual isotopic variation in zooplankton (e.g. cladoceran δ13C annual range = 25.6�), it may be impossible to definitively assign consumers from small, eutrophic stratified lakes to pelagic or benthic-littoral food webs. We present evidence that some components of the Plu�see food web (large bream) may be subsidised by carbon of methanogenic origin.

Digital Exposure of English Place-names (DEEP)

Place-names are a fundamental concept in all academic collections: everything happens somewhere. Contemporary place-names are comprehensively represented in digital gazetteer and geospatial web services such as GeoNames. However, despite millions of pounds of investment by JISC and other agencies in historical online resources in recent years, there is currently no equivalent for historic place-names. This project will digitize the entire 86 volume corpus of the Survey of English Place-Names (SEPN), the ultimate authority on historic place-names in England, and make its 4 million forms available.