Since the previous edition of this collection, the World Wide Web has unequivocally laid any lingering questions regarding its longevity and global impact to rest. Several multi-Billion-user services including Google and Facebook have become central to modern life in the first world, while Internetand Web-related technology has permeated both business and personal interactions. The Web is undoubtedly here to stayat least for the foreseeable future.

But search engines often run on a single web snapshot and hence are not able to track temporal changes on a website. Furthermore the ranking algorithms on search engines, as Google, only refer to the number of in-links a website has without considering how fresh, means how new created or updated, the content is. This leads to the fact that older websites, which have more time to gather in-links, rank higher than newer (but fresher) websites with fewer in-links.

In this paper, we discuss the architecture and implementation of a Semantic Web search engine, appropriately called the Semantic Web Search Engine (SWSE). Following traditional search engine architecture, SWSE consists of crawling, data enhancing, indexing and a user interface for search, browsing and retrieval of information; unlike traditional search engines, SWSE operates over RDF Web data—loosely also known as Linked Data—which implies unique challenges for the system design, architecture, algorithms, implementation and user interface. In particular, many challenges exist in adopting Semantic Web technologies for Web data: the unique challenges of the Web—in terms of scale, unreliability, inconsistency and noise—are largely overlooked by the current Semantic Web standards. In this paper, we detail the current SWSE system, initially detailing the architecture and later elaborating upon the function, design, implementation and performance of each individual component. In so doing, we also give an insight into how current Semantic Web standards can be tailored, in a best-effort manner, for use on Web data. Throughout, we offer evaluation and complementary argumentation to support our design choices, and also offer discussion on future directions and open research questions. Later, we also provide candid discussion relating to the difficulties currently faced in bringing such a search engine into the mainstream, and lessons learnt from roughly five years working on the Semantic Web Search Engine project.

Many daily activities present information in the form of a stream of text, and often people can benefit from additional information on the topic discussed. TV broadcast news can be treated as one such stream of text; in this paper we discuss finding news articles on the web that are relevant to news currently being broadcast.We evaluated a variety of algorithms for this problem, looking at the impact of inverse document frequency, stemming, compounds, history, and query length on the relevance and coverage of news articles returned in real time during a broadcast. We also evaluated several postprocessing techniques for improving the precision, including reranking using additional terms, reranking by document similarity, and filtering on document similarity. For the best algorithm, 84%-91% of the articles found were relevant, with at least 64% of the articles being on the exact topic of the broadcast. In addition, a relevant article was found for at least 70% of the topics.

Mining for association rules in market basket data has proved a fruitful area of research. Measures such as conditional probability (confidence) and correlation have been used to infer rules of the form “the existence of item A implies the existence of item B.” However, such rules indicate only a statistical relationship between A and B. They do not specify the nature of the relationship: whether the presence of A causes the presence of B, or the converse, or some other attribute or phenomenon causes both to appear together. In applications, knowing such causal relationships is extremely useful for enhancing understanding and effecting change. While distinguishing causality from correlation is a truly difficult problem, recent work in statistics and Bayesian learning provide some avenues of attack. In these fields, the goal has generally been to learn complete causal models, which are essentially impossible to learn in large-scale data mining applications with a large number of variables. In this paper, we consider the problem of determining casual relationships, instead of mere associations, when mining market basket data. We identify some problems with the direct application of Bayesian learning ideas to mining large databases, concerning both the scalability of algorithms and the appropriateness of the statistical techniques, and introduce some initial ideas for dealing with these problems. We present experimental results from applying our algorithms on several large, real-world data sets. The results indicate that the approach proposed here is both computationally feasible and successful in identifying interesting causal structures. An interesting outcome is that it is perhaps easier to infer the lack of causality than to infer causality, information that is useful in preventing erroneous decision making.

The importance of a Web page is an inherently subjective matter, which depends on the readers interests, knowledge and attitudes. But there is still much that can be said objectively about the relative importance of Web pages. This paper describes PageRank, a mathod for rating Web pages objectively and mechanically, effectively measuring the human interest and attention devoted to them. We compare PageRank to an idealized random Web surfer. We show how to efficiently compute PageRank for large numbers of pages. And, we show how to apply PageRank to search and to user navigation.

The World Wide Web is a vast resource for information. At the same time it is extremely distributed. A particular type of data such as restaurant lists may be scattered across thousands of independent information sources in many different formats. In this paper, we consider the problem of extracting a relation for such a data type from all of these sources automatically. We present a technique which exploits the duality between sets of patterns and relations to grow the target relation starting from a small sample. To test our technique we use it to extract a relation of (author;title) pairs from the World Wide Web.

A great challenge for data mining techniques is the huge space of potential rules which can be generated. If there are tens of thousands of items, then potential rules involving three items number in the trillions. Traditional data mining techniques rely on downward-closed measures such as support to prune the space of rules. However, in many applications, such pruning techniques either do not sufficiently reduce the space of rules, or they are overly restrictive. We propose a new solution to this problem, called Dynamic Data Mining (DDM). DDM foregoes the completeness offered by traditional techniques based on downward-closed measures in favor of the ability to drill deep into the space of rules and provide the user with a better view of the structure present in a data set. Instead of a single deterministic run, DDM runs continuously, exploring more and more of the rule space. Instead of using a downward-closed measure such as support to guide its exploration, DDM uses a user-defined measure called weight, which is not restricted to be downward closed. The exploration is guided by a heuristic called the Heavy Edge Property. The system incorporates user feedback by allowing weight to be redefined dynamically. We test the system on a particularlly difficult data set - the word usage in a large subset of the World Wide Web. We find that Dynamic Data Mining is an effective tool for mining such difficult data sets.

In this paper, we present Google, a prototype of a large-scale search engine which makes heavy use of the structure present in hypertext. Google is designed to crawl and index the Web efficiently and produce much more satisfying search results than existing systems. The prototype with a full text and hyperlink database of at least 24million pages is available at http://google.stanford.edu/

The amount of information available online has grown enormously over the past decade. Fortunately, computing power, disk capacity, and network bandwidth have also increased dramatically. It is currently possible for a university research project to store and process the entire World Wide Web. Since there is a limit on how much text humans can generate, it is plausible that within a few decades one will be able to store and process all the human-generated text on the Web in a shirt pocket. The Web is a very rich and interesting data source. In this paper, we describe the Stanford WebBase, a local repository of a significant portion of the Web. Furthermore, we describe a number of recent experiments that leverage the size and the diversity of the WebBase. First, we have largely automated the process of extracting a sizable relation of books (title, author pairs) from hundreds of data sources spread across the World Wide Web using a technique we call Dual Iterative Pattern Relation Extraction. Second, we have developed a global ranking of Web pages called PageRank based on the link structure of the Web that has properties that are useful for search and navigation. Third, we have used PageRank to develop a novel search engine called Google, which also makes heavy use of anchor text. All of these experiments rely significantly on the size and diversity of the WebBase.

One of the most well-studied problems in data mining is mining for association rules in market basket data. Association rules, whose significance is measured via support and confidence, are intended to identify rules of the type, “A customer purchasing item A often also purchases item B.” Motivated by the goal of generalizing beyond market baskets and the association rules used with them, we develop the notion of mining rules that identify correlations (generalizing associations), and we consider both the absence and presence of items as a basis for generating rules. We propose measuring significance of associations via the chi-squared test for correlation from classical statistics. This leads to a measure that is upward closed in the itemset lattice, enabling us to reduce the mining problem to the search for a border between correlated and uncorrelated itemsets in the lattice. We develop pruning strategies and devise an efficient algorithm for the resulting problem. We demonstrate its effectiveness by testing it on census data and finding term dependence in a corpus of text documents, as well as on synthetic data.

We consider the problem of analyzing market-basket data and present several important contributions. First, we present a new algorithm for finding large itemsets which uses fewer passes over the data than classic algorithms, and yet uses fewer candidate itemsets than methods based on sampling. We investigate the idea of item reordering, which can improve the low-level efficiency of the algorithm. Second, we present a new way of generating “implication rules,” which are normalized based on both the antecedent and the consequent and are truly implications (not simply a measure of co-occurrence), and we show how they produce more intuitive results than other methods. Finally, we show how different characteristics of real data, as opposed by synthetic data, can dramatically affect the performance of the system and the form of the results.

In a digital library system, documents are available in digital form and therefore are more easily copied and their copyrights are more easily violated. This is a very serious problem, as it discourages owners of valuable information from sharing it with authorized users. There are two main philosophies for addressing this problem: prevention and detection. The former actually makes unauthorized use of documents difficult or impossible while the latter makes it easier to discover such activity.In this paper we propose a system for registering documents and then detecting copies, either complete copies or partial copies. We describe algorithms for such detection, and metrics required for evaluating detection mechanisms (covering accuracy, efficiency, and security). We also describe a working prototype, called COPS, describe implementation issues, and present experimental results that suggest the proper settings for copy detection parameters.

Given user data, one often wants to find approximate matches in a large database. A good example of such a task is finding images similar to a given image in a large collection of images. We focus on the important and technically diffcult case where each data element is high dimensional, or more generally, is represented by a point in a large metric spaceand distance calculations are computationally expensive. In this paper we introduce a data structure to solve this problem called a GNAT { Geometric Near-neighbor Access Tree. It is based on the philosophy that the data structure should act as a hierarchical geometrical model of the data as opposed to a simple decomposition of the data that does not use its intrinsic geometry. In experiments, we find that GNAT's outperform previous data structures in a number of applications. Keywords { near neighbor, metric space, approximate queries, data mining, Dirichlet domains, Voronoi regions