Chapter 2 BACKGROUND OF WEB MINING

Transcription

1 Chapter 2 BACKGROUND OF WEB MINING Overview 2.1. Introduction to Data Mining Data mining is an important and fast developing area in web mining where already a lot of research has been done. Recently, many organizations got aware of its potentials, especially for applications in marketing, science and finance. However, a structured methodology is a crucial requirement for a successful practical application of data mining [61]. Data are any facts, numbers, images, or text that can be processed by a computer. Today, organizations are accumulating vast and growing amounts of data in different formats. Data mining is defined as the process of analyzing data from different perspectives and summarizing it into useful information - information that can be used to increase revenue, cuts costs, or some times both. It is a powerful new technology with great potential helping organizations/companies focus on the most important information in their data warehouses Data mining tools usage Data mining tools predict user behaviors and future trends, allowing businesses to make knowledge-driven and proactive decisions. The automated, prospective analyses offered by data mining move beyond the analyses of past 6

2 events provided by retrospective tools. Data mining tools can provide business solutions that traditionally were time consuming. They scour the whole database for hidden forms, finding predictive information that experts usually miss because of the huge content in the database which lies outside their expectations Data mining software Data mining software is one of the analytical tools used for data analyses. It allows users to analyze data from many different dimensions, categorize it, and summarize accordingly. Technically, data mining is the process of finding out the correlations among many numbers of fields in huge relational databases Data Mining Types Spatial Data Mining The main difference between data mining in relational DBS and in spatial DBS is that attributes of the neighbors of some object of interest may have an influence on the object and therefore have to be considered as well. The Precisely and clearly expressed location and extension of spatial objects define not so directly expressed relations of spatial neighborhood (such as topological, distance and direction) that are used by spatial data mining algorithms. Therefore, new techniques are required for efficient and effective data mining. Spatial trends describe a regular change of non-spatial attributes while moving away from certain start objects. Global and local trends can be distinguished. To detect and explain such spatial trends, e.g. with respect to the economic power, is an important issue in economic geography. 7

3 Database Primitives for Spatial Data Mining Many tools are developed for mining in spatial databases that are sufficient to express most of the algorithms for spatial data mining and they can be very efficiently get supported by DBMS. The use of such database primitives will enable the integration of spatial data mining with existing DBMS s and will speed-up the development of new spatial data mining algorithms. The database primitives are based on the concepts of neighborhood graphs and paths Algorithms for Spatial Data Mining Many new algorithms for spatial characterization and spatial trend analysis are developed today. For spatial characterization it is important that the class membership of a database object is determined not only by its non spatial attributes but also by the attributes of objects in its neighborhood. In spatial trend analysis, patterns of change of some non-spatial attributes in the neighborhood of a database object can also be determined Text Mining Text Mining is the discovery by computer of new, previously unknown information, by automatically extracting information from different written resources. A key element is the link; these link together the extracted information to form new facts that are to be explored further by more conventional means of experimentation[68]. Text mining, in short discovers new pieces of knowledge, from approaches that find overall trends in textual data. Many organizations are using the output of 8

4 such programs to try to link together information in interesting ways. For example, one can extract all the names of people and companies that occur in news text surrounding the topic of wireless technology to try to infer the players in that field. There are a number of companies that are investigating on this kind of application. Text mining is more than what we're familiar with in web search. While searching, user is typically looking for something that is already known and has been written by someone else. The problem is ignoring all the material that currently isn't relevant to our needs in order to find the relevant information. In text mining, the goal is to discover the unknown information, something that no one yet knows and not written by anyone. Text mining is a variation on data mining field that tries to find interesting patterns from the huge databases. An example in data mining is using consumerpurchasing patterns to predict which products to place together on shelves, or to offer coupons on some products, and so on. For example, if we buy a camera, we often buy batteries along with it. A related application is automatic detection of deception, such as in credit cards usage. Analysts go through huge numbers of credit card records to find out deviations from the usual spending patterns. A typical example is the use of a credit card to buy a small amount of food followed by an international flight ticket. The claim is that the first purchase makes sure that the card is active. The difference between regular data mining and text mining is that in text mining the patterns are extracted from natural language text rather than from 9

5 structured databases. These databases are designed for programs to process automatically. There are programs that can extract information from text with reasonable accuracy in a regularized structure. For instance, programs that can read resumes and extract the candidate's name, address, job skills, and so on, with 80% accuracy. There are certain fundamental limitations of text mining: that we will not be able to write programs that fully interpret text for a very long time, and second, that the information one needs is often not recorded in textual form Web Mining As many believe, it is Etzioni first proposed the term of Web mining in his paper [24] in In this paper, he claimed the Web mining is the use of data mining techniques to automatically discover and extract information from World Wide Web documents and services. Many of the following researchers cited this explanation in their works. In the same paper, Etzioni came up with the question: Whether effective Web mining is feasible in practice? Today, with the tremendous growth of the data sources available on the Web and the dramatic popularity of e- commerce in the business community, Web mining has become the focus of quite a few research projects, papers and some commercial products. In Etzioni [24],Khasawneh[63] and Kosala and Blocked [34],Liu[64] they suggested a similar way to decompose Web mining into the following subtasks: a. Resource Discovery: the task of retrieving the intended information from Web. 10

6 b. Information Extraction: automatically selecting and preprocessing specific information from the retrieved Web resources. c. Generalization: automatically discovers general patters at the both individual Web sites and across multiple sites. d. Analysis: analyzing the mined pattern. In brief, Web mining is a technique to discover and analyze the useful information from the Web data. Madriaet al. [36] claim the Web involves three types of data: data on the Web (content), Web log data (usage) and Web structure data. Cooley [6] classified the data type as content data, structure data, usage data, and user profile data. Spiliopoulou [18],Gaber[70] categorized the Web mining into Web usage mining, Web text mining and user modeling mining; while today the most recognized categories of the Web data mining are Web content mining, Web structure mining, and Web usage mining according to Borges and Levene [12] and Kosala and Blocked [34]. It is clear that the classification as seen Figure 2.1 is based on what type of Web data to mine. ii

7 WebMining Taxonomy 1 WebM Mining i WebSIructure Mining Wei) Mining ZIP _ Web Usage lining MM drniiim SiteMIMft System topwt PmlM MM} inifl} Usage Hlig SjfM Fig 2.1. Web Mining Taxonomy I. Web Content Mining Web content mining describes the automatic search of information resource available online, and involves mining web data contents. In the Web mining domain, Web content mining essentially is an analog of data mining techniques for relational databases, since it is possible to find similar types of knowledge from the unstructured data residing in Web documents. The Web document usually contains several types of data, such as text, image, audio, video, metadata and hyperlinks. Some of them are semi structured such as HTML documents, or a more structured data like the data in the tables or database generated HTML pages, but most of the data is unstructured text data. The unstructured characteristic of Web data forces the Web content mining towards a more complicated approach. 12

8 The Web content mining is differentiated from two different points of view, Cooley et al. [32],Chen[69]: Information Retrieval View and Database View. Kosala and Blocked [34] summarized the research works done for unstructured data and semi structured data from information retrieval view. It shows that most of the researches use bag of words, which is Web Mining Taxonomy based on the statistics about single words in isolation, to represent unstructured text and take single word found in the training corpus as features. For the semi-structured data, all the works use the HTML structures inside the documents and some use the hyperlink structure between the documents for document representation. As for the database view, in order to have the better information management and querying on the Web, the mining always tries to infer the structure of the Web site of to transform a Web site to become a database. Chakrabarti [35] provides an in-depth survey of the research on the application of the techniques from machine learning, statistical pattern recognition, and data mining to analyzing hypertext. It is a good resource to be aware of the recent advances in content mining research. Multimedia data mining is part of the content mining, which is engaged to mine the high-level information and knowledge from large online multimedia sources. Multimedia data mining on the Web has gained many researchers attention recently. Working towards a unifying framework for representation, problem solving, and learning from multimedia is really a challenge, this research area is still in its infancy indeed, and many works are waiting to be done. The details about multimedia mining are given in Zaiane et al. [22]. 13

9 Web Structure Mining Most of the Web information retrieval tools only use the textual information, while ignoring the link information that could be very valuable. The goal of Web structure mining is to generate structural summary about the Web site and Web page. Technically, Web content mining mainly focuses on the structure of inner document, while Web structure mining tries to discover the link structure of the hyperlinks at the interdocument level. Based on the topology of the hyperlinks, Web structure mining will categorize the Web pages and generate the information, such as the similarity and relationship between different Web sites. Web structure mining can also have another direction discovering the structure of Web document itself. This type of structure mining can be used to reveal the structure (schema) of Web pages; this would be good for navigation purpose and make it possible to compare/integrate Web page schemes. This type of structure mining will facilitate introducing database techniques for accessing information in Web pages by providing a reference schema. What is the structural information, and how to discover it? Madria et al. [36] gave a detailed description about how to discover interesting and informative facts describing the connectivity in the Web subset, based on the given collection of interconnected web documents. The structural information generated from the Web structure mining includes the follows: the information measuring the frequency of the local links in the Web tuples in a Web table; the information measuring the frequency of Web tuples in a Web table containing links that are interior and the links that are within the same document; the information measuring the frequency of Web tuples in a Web table that contains links that are global and the links that 14

10 span different Web sites; the information measuring the frequency of identical Web tuples that appear in a Web table or among the Web tables. In general, if a Web page is linked to another Web page directly, or the Web pages are neighbors, we would like to discover the relationships among those Web pages. The relations maybe fall in one of the types, such as they related by synonyms or ontology, they may have similar contents; both of them may sit in the same Web server therefore created by the same person. Another task of Web structure mining is to discover the nature of the hierarchy or network of hyperlinks in the Web sites of a particular domain. This may help to generalize the flow of information in Web sites that may represent some particular domain; therefore, the query processing will be easier and more efficient. Web structure mining has a nature relation with the Web content mining, since it is very likely that the Web documents contain links, and they both use the real or primary' data on the Web. It is quite often to combine these two mining tasks in an application Web Usage Mining Web usage mining tries to discover the useful information from the secondary data derived from the interactions of the users while surfing on the Web. Web Usage mining studies can be classified under the headlines of. General Web Usage Mining, Site Modification, System Improvement and Personalization. General Web Usage Mining systems aim to discover general trends and patterns from the log files either by adapting well known data mining techniques or by proposing new data mining techniques. 15

11 The objective of the Site Modification systems is to improve the design of a web site by suggesting modifications in its content and structure. The research on System Improvement focuses on using thb web usage mining for improving the web traffic. Lastly, personalization systems aim to understand individual trends used for personalizing the web sites. Throughout the following sections, detailed description of the projects belonging to each category is given Pattern Discovery Techniques Each Web mining process requires a pattern discovery phase in which the algorithms and techniques from several research areas, such as data mining, machine learning, statistics, and pattern recognition techniques can be adopted. Throughout this section, some of the techniques used for pattern discovery are explained Statistical Analysis Statistical techniques are the most powerful tools in extracting knowledge about visitors to a Web site. The analysts may perform different kinds of descriptive statistical analyses based on different variables when analyzing the session file. By analyzing the statistical information contained in the periodic Web system report, the extracted report can be potentially useful for improving the system performance, enhancing the security of the system, facilitation the site modification task, and providing support for marketing decisions, Cooley [6]. 16

12 Association Rules In the Web domain, the pages, which are most often referenced together, can be put in one single server session by applying the association rule generation. Association rule mining techniques can be used to discover unordered correlation between items found in a database of transactions. Cooley [6] pointed that in the term of the Web usage mining, the association rules refer to sets of pages that are accessed together with a support value exceeding some specified threshold. The support is the percentage of the transactions that contain a given pattern. The Web designers can restructure their Web sites efficiently with the help of the presence or absence of the association rules. When loading a page from a remote site, association rules can be used as a trigger for pre fetching documents to reduce user perceived latency Clustering Clustering analysis is a technique to group together users or data items (pages) with the similar characteristics. Clustering of user information or pages can facilitate the development and execution of future marketing strategies, Cooley [6]. Clustering of users will help to discover the group of users, who have similar navigation pattern. It is very useful for inferring user demographics to perform market segmentation in E-commerce applications or provide personalized Web content to the individual users. The clustering of pages is useful for Internet search engines and Web service providers, since it can be used to discover the groups of pages having related content. 17

13 Classification Classification is the technique to map a data item into one of several predefined classes. In the Web domain, Web master or marketer will have to use this technique if he/she wants to establish a profile of users belonging to a particular class or category. This requires extraction and selection of features that best describe the properties of a given class or category. Cooley [6] indicates that the classification can be done by using supervised inductive learning algorithms such as decision tree classifiers, naive Bayesian classifiers, k-nearest neighbor classifier, Support Vector Machines etc Sequential Pattern This technique intends to find the inter-session pattern, such that a set of the items follows the presence of another in a time ordered set of sessions or episodes. It is very meaningful for the Web marketer to predict the future trend, which help to place advertisements aimed at certain user groups. Sequential patterns also include some other types of temporal analysis such as trend analysis, change point detection, or similarity analysis, Cooley [6] Dependency Modeling The goal of this technique is to establish a model that is able to represent significant dependencies among the various variables in the Web domain. The modeling technique provides a theoretical framework for analyzing the behavior of users, and is potentially useful for predicting future Web resource consumption. 18

14 Web Usage Mining Applications Web usage mining is the application used in data mining to analyze and discover required hidden patterns of user s usage data on the web. The usage data records the user s interactions when the user browses or makes transactions on the web site. An activity involves the automatic discovery of patterns from one or more Web servers. Organizations often generate and collect large volumes of data; most of this information is usually generated automatically by Web servers and collected in the server log. Analyzing such data can help these organizations to determine the importance/value of particular customers, cross marketing strategies across products and the effectiveness of promotional campaigns, etc. The first web analysis tools were simply providing mechanisms to report user behavior as recorded in the servers. Using such tools, it was possible to determine such information as the number of accesses to the server, the times or time intervals of visits as well as the domain names and the URLs of users of the Web server. However, generally, these tools do not provide the analysis of data relationships among the accessed directories and files within the Web space. Now many sophisticated techniques for analysis and discovery of patterns are emerging. They fall into two categories: Pattern Discovery Tools and Pattern Analysis Tools. Another important application of Web Usage Mining is Web Link recommendation. One of the last trends is represented by the online monitoring of page accesses to render personalized pages on the basis of similar visit patterns. Web usage mining is the application of data mining techniques to discover the patterns of usage from the Web data, in order to understand and serve better to the needs of Web-based applications. Web usage mining comprises of three phases, 19

15 namely preprocessing, pattern discovery, and pattern analysis. Both research and practice communities have increased their interest rapidly in Web usage mining within the given application potential. Personalizing the web experience for a user is the prolonged endeavor of many Web-based applications. 20