Data-mining

Scientific insti uments can easily pi oduce and collect huge amounts of data. Nevertheless, it is seldom possible to read the relevant information in these data directly. Therefore more elaborate method. have to be applied for information extraction. 

Data mining provides methods foi the exti action of implicit oi hidden information from large data sets and comprises procedures for the generation of reasonable and dependable secondai information. 

Data Mining is the core of the more comprehensive process of knowledge dis-coveiy in data bases (KDD). However, the term data mining is often used synonymously with KDD. KDD describes the process of extracting and storing data and also includes methods for data preparation such as data cleaning, data selection, and data transformation as well as evaluation, presentation, and visualization of the results after the data mining process. 

The methods used for data mining also comprise those methods that have already been described in Sections 9,2-9,7. 

This section will therefore focus on the aims and tasks of data mining and refer to the methods where applicable. A thorough description of data mining is given in Ref. [20]. 

The US Constitution, federal statutes and regulations, and state law combine to govern the collection, use, and disclosure of information. The Constitution provides certain privacy protections, but does not explicitly protect information privacy. Generally, federal law addresses privacy issues and personal information by topic (e.g., education, telecommunications, privacy, health information, motor vehicle, communications and communications records, financial and credit information, children s online (Internet) privacy) The individual s interests are usually balanced with the government s need, with authorization for personal information normally being sought through warrants, subpoenas, and court orders [120]. 

---

To understand visual data mining and information visualization techniques... 

The area of machine learning is thus quite broad, and different people have different notions about the domain of machine learning and what kind of techniques belong to this field. We will meet a similar problem of defining an area and the techniques involved in the field of "data mining , as discussed in Section 9.8. We will use the term "machine learning in this chapter to collect aU the methods that involve learning from data. 

It extends the usage of statistical methods and combines it with machine learning methods and the application of expert systems. The visualization of the results of data mining is an important task as it facilitates an interpretation of the results. Figure 9-32 plots the different disciplines which contribute to data mining. 

Data mining can fulfill various different tasks such as classification, clustering and similarity detection, prediction, estimation, or description retrieval, which are described in Sections 9.8.1-9.8.5. 

Machine Learning ------M Data Mining [<-------- Expert Systems... 

A very important data mining task is the discovery of characteristic descriptions for subsets of data, which characterize its members and distinguish it from other subsets. Descriptions can, for example, be the output of statistical methods like average or variance. 

Data mining in chemistry focuses on the extraction and evaluation of information in chemical data sets. In contrast to other fields of data mining applications, chemical data mining does not confine itself to conventional database queries but rather generates new information from the data. 

Higher quality of the resulting patterns The natural capabiUty of human beings to visually recognize patterns and relations can be used and leads to a more effective data mining process. 

Increased trust in pattern recognition The active user involvement in the data mining process can lead to a deeper understanding of the data and increases the trust in the resulting patterns. In contrast, "black box" systems often lead to a higher uncertainty, because the user usually does not know, in detail, what happened during the data analysis process. This may lead to a more difficult data interpretation and/or model prediction. 

Handling of complex data sets Visual data mining methods especially show huge advantages over classical approaches if only Httle information about the data is known or if the expected patterns and relationships are not clearly defined. Furthermore, very inhomogeneous data sets or data with a high noise level can still be analyzed by these methods. 

The explorative analysis of data sets by visual data mining applications takes place in a three-step process During the first step (overview), the user can obtain an overview of the data and maybe can identify some basic relationships between specific data points. In the second step (filtering), dynamic and interactive navigation, selection, and query tools will be used to reorganize and filter the data set. Each interaction by the user will lead to an immediate update of the data scene and will reveal the hidden patterns and relationships. Finally, the patterns or data points can be analyzed in detail with specific detail tools. 

Visual data mining allows the visualization and detection of hidden relationships in sets of data. 

U.M. Fayyad, G. Piatetsky-Shapiro, P. Smyth, R. Uthurusamy (Eds.), Advances in Knowledge Discovery and Data Mining, AAAI/MIT Press, Menlo Park, 1996. 

B. Thuraisingham, Data Mining Technologies, Techniques, Tools, and Trends, CRC Press, Boca Raton, 1999. 

U. Fayyad, G. Grinstein, A. Wierse, Information Visualization in Data Mining and Knowledge Discovery, Morgan Kaufman Publishers, San Frandsco, USA, 2002. 

D. Keim, Information Visualization and Visual Data Mining, IEEE Trans. Visualization and Computer Graphics, 2002, 8,100-107. 

A most important task in the handling of molecular data is the evaluation of "hidden information in large chemical data sets. One of the differences between data mining techniques and conventional database queries is the generation of new data that are used subsequently to characterize molecular features in a more general way. Generally, it is not possible to hold all the potentially important information in a data set of chemical structures. Thus, the extraction of relevant information and the production of reliable secondary information are important topics. 

A variety of methods have been developed by mathematicians and computer scientists to address this task, which has become known as data mining (see Chapter 9, Section 9.8). Fayyad defined and described the term data mining as the nontrivial extraction of impHcit, previously unknown and potentially useful information from data, or the search for relationships and global patterns that exist in databases [16]. In order to extract information from huge quantities of data and to gain knowledge from this information, the analysis and exploration have to be performed by automatic or semi-automatic methods. Methods applicable for data analysis are presented in Chapter 9. 

Nowadays a broad range of methods is available in the field of chemoinfor-matics. These methods will have a growing impact on drug design. In particular, the discovery of new lead structures and their optimization will profit by virtual saeening [17, 66, 100-103]. The huge amounts of data produced by HTS and combinatorial chemistry enforce the use of database and data mining techniques. 

P. Smyth, From data mining to knowledge discovery An overview, in Advances in Knowledge Discovery and Datamining, U. M. Fayyad,... 

---