Mining information and knowledge from large databases has been recognized by many researchers as a key research topic in database systems and machine learning, and by many industrial companies as an important area with an opportunity of major revenues. Researchers in many different fields have shown great interest in data mining. Several emerging applications in information providing services, such as data warehousing and on-line services over the Internet, also call for various data mining techniques to better understand user behavior, to improve the service provided, and to increase the business opportunities. In response to such a demand, this article is to provide a survey, from a database researcher's point of view, on the data mining techniques developed recently. A classification of the available data mining techniques is provided and a comparative study of such techniques is presented.

Knowledge discovery facilitates querying database knowledge and intelligent query answering in database systems. In this paper, we investigate the application of discovered knowledge, concept hierarchies, and knowledge discovery tools for intelligent query answering in database systems. A knowledge-rich data model is constructed to incorporate discovered knowledge and knowledge discovery tools. Queries are classified into data queries and knowledge queries. Both types of queries can be answered directly by simple retrieval or intelligently by analyzing the intent of query and providing generalized, neighborhood or associated information using stored or discovered knowledge. Techniques have been developed for intelligent query answering using discovered knowledge and/or knowledge discovery tools, which includes generalization, data summarization, concept clustering, rule discovery, query rewriting, deduction, lazy evaluation, application of multiplelayered databases, etc. Our study shows that knowledge discovery substantially broadens the spectrum of intelligent query answering and may have deep implications on query answering in data- and knowledge-base systems.

Data mining is the discovery of knowledge and useful information from the large amounts of data stored in databases. With the increasing popularity of object-oriented database systems in advanced database applications, it is important to study the data mining methods for object-oriented databases because mining knowledge from such databases may improve understanding, organization, and utilization of the data stored there.

Recent advances in the development of the INLEN system for multistrategy data exploration are briefly reviewed. These advances include the development of a meta-level language for data mining and knowledge discovery, called knowledge generation language (KGL), and the employment of a new type of attributes, called structured attributes. These features are illustrated by an example concerned with determining economic and demographic patterns in a database containing facts about the countries of the world. The results demonstrate a high utility of INLEN for data mining and knowledge discovery. Introduction The availability of very large volumes of data in the electronic form has created a problem of deriving from them useful, task-oriented knowledge. Traditional data analysis techniques, which include statistical and numerical methods, are oriented primarily toward the extraction of quantitative data characteristics, and as such have inherent limitations. For example, statistical techn...

Two issues arising in the design of trace-file based performance analysis tools are discussed: handling categorical rather than just numeric data and correctly inferring system behavior from trace data by using not one but multiple trace files. The issues are illustrated using the problem of determining whether damaging oscillations occur in the all points shortest path algorithm when used for routing messages between processors. Solutions used by the Chitra trace analysis tool are discussed.

Attribute-oriented induction is a set-oriented database mining method which generalizes the task-relevant subset of data attribute-by-attribute, compresses it into a generalized relation, and extracts from it the general features of data. In this chapter, the power of attribute-oriented induction is explored for the extraction from relational databases of different kinds of patterns, including characteristic rules, discriminant rules, cluster description rules, and multiple-level association rules. Furthermore, it is shown that the method is efficient, robust, with wide applications, and extensible to knowledge discovery in advanced database systems, including object-oriented, deductive, and spatial database systems. The implementation status of DB- MINER a system prototype which applies the method, is also reported here.

Time is a common factor in most of the large databases. The time factor in such data can be utilized during the knowledge discovery process to overcome some limitations, such as the size of the data, of many learning systems. The paper presents an experiment for discovering knowledge in large temporal databases. The method divides the learning space into subspaces each corresponds to one time interval. The process of discovery is performed on the data available in each subspace, for a given set of decision classes, using the learning system AQ15. The method determines a set of attributes relevant to the discovery process using the important score (IS) system. The experiment is done on international TRADE data; the data is concerned with the imports and exports between the US and the world. The preliminary results show that strong relationships in the original data can be discovered over different subsets of the data.