Data Mining Mehmed Kantardzic (good english books to read .txt) 📖

The SOM is one efficient solution for the problems of unstructured visualization of documents and unstructured data. With a properly constructed SOM, you can analyze literally millions of unstructured documents that can be merged into a single SOM. The SOM deals not only with individual unstructured documents but relationships between documents as well. The SOM may show text that is correlated to other text. For example, in the medical field, working with medical patient records, this ability to correlate is very attractive. The SOM also allows the analyst to see the larger picture as well as drilling down to the detailed picture. The SOM goes down to the individual stemmed-text level, and that is as accurate as textual processing can become. All these characteristics have resulted in the growing popularity of SOM visualizations in order to assist visual inspection of complex high-dimensional data. For the end user the flexibility of the SOM algorithm is defined through a number of parameters. For appropriate configuration of the network, and tuning the visualization output, user-defined parameters include grid dimensions (2-D, 3-D), grid shape (rectangle, hexagon), number of output nodes, neighborhood function, heighborhood size, learning-rate function, initial weights in the network, way of learning and number of iterations, and order of input samples.

15.6 VISUALIZATION SYSTEMS FOR DATA MINING

Many organizations, particularly within the business community, have made significant investments in collecting, storing, and converting business information into results that can be used. Unfortunately, typical implementations of business “intelligence software” have proven to be too complex for most users except for their core reporting and charting capabilities. Users’ demands for multidimensional analysis, finer data granularity, and multiple-data sources, simultaneously, all at Internet speed, require too much specialist intervention for broad utilization. The result is a report explosion in which literally hundreds of predefined reports are generated and pushed throughout the organization. Every report produces another. Presentations get more complex. Data are exploding. The best opportunities and the most important decisions are often the hardest to see. This is in direct conflict with the needs of frontline decision makers and knowledge workers who are demanding to be included in the analytical process.

Presenting information visually, in an environment that encourages the exploration of linked events, leads to deeper insights and more results that can be acted upon. Over the past decade, research on information visualization has focused on developing specific visualization techniques. An essential task for the next period is to integrate these techniques into a larger system that supports work with information in an interactive way, through the three basic components: foraging the data, thinking about data, and acting on data.

The vision of a visual data-mining system stems from the following principles: simplicity, visibility, user autonomy, reliability, reusability, availability, and security. A visual data-mining system must be syntactically simple to be useful. Simple does not mean trivial or non-powerful. Simple to learn means use of intuitive and friendly input mechanisms as well as instinctive and easy-to-interpret output knowledge. Simple to apply means an effective discourse between humans and information. Simple to retrieve or recall means a customized data structure that facilitates fast and reliable searches. Simple to execute means a minimum number of steps needed to achieve the results. In short, simple means the smallest, functionally sufficient system possible.

A genuinely visual data-mining system must not impose knowledge on its users, but instead guide them through the mining process to draw conclusions. Users should study the visual abstractions and gain insight instead of accepting an automated decision. A key capability in visual analysis, called visibility, is the ability to focus on particular regions of interest. There are two aspects of visibility: excluding and restoring data. The exclude process eliminates the unwanted data items from the display so that only the selected set is visible. The restore process brings all data back, making them visible again.

A reliable data-mining system must provide for estimated error or accuracy of the projected information in each step of the mining process. This error information can compensate for the deficiency that an imprecise analysis of data visualization can cause. A reusable, visual, data-mining system must be adaptable to a variety of environments to reduce the customization effort, provide assured performance, and improve system portability. A practical, visual, data-mining system must be generally and widely available. The quest for new knowledge or deeper insights into existing knowledge cannot be planned. It requires that the knowledge received from one domain adapt to another domain through physical means or electronic connections. A complete, visual, data-mining system must include security measures to protect the data, the newly discovered knowledge, and the user’s identity because of various social issues.

Through data visualization we want to understand or get an overview of the whole or a part of the n-dimensional data, analyzing also some specific cases. Visualization of multidimensional data helps decision makers to

1. slice information into multiple dimensions and present information at various levels of granularity,

2. view trends and develop historical tracers to show operations over time,

3. produce pointers to synergies across multiple dimensions,

4. provide exception analysis and identify isolated (needle in the haystack) opportunities,

5. monitor adversarial capabilities and developments,

6. create indicators of duplicative efforts,

7. conduct What-If Analysis and Cross-Analysis of variables in a data set.

Visualization tools transform raw experimental or simulated data into a form suitable for human understanding. Representations can take on many different forms, depending on the nature of the original data and the information that is to be extracted. However, the visualization process that should be supported by modern, visualization-software tools can generally be subdivided into three main stages: data preprocessing, visualization mapping, and rendering. Through these three steps the tool has to answer the questions: What should be shown in a plot? How should one work with individual plots? How should multiple plots be organized?

Data preprocessing involves such diverse operations as interpolating irregular data, filtering and smoothing raw data, and deriving functions for measured or simulated quantities. Visualization mapping is the most crucial stage of the process, involving design and adequate representation of