Concept Data Analysis : Theory and Applications
Claudio Carpineto, Giovanni Romano
立即出貨 (庫存 < 3)
The advent of the Web, along with the unprecedented amount of data available in electronic format, has dramatically increased the need for tools that support the users in retrieving, understanding and mining the information and knowledge contained in such data.
Concept data analysis differs from statistical data analysis in that the emphasis is on recognising and generalising the structure of symbolic data through a mathematical representation termed a concept lattice. Thanks to its simplicity, elegance and versatility, concept data analysis can effectively support various kinds of content management tasks using different or heterogeneous types of data.
- Provides a comprehensive treatment of the full range of techniques developed for concept data analysis covering creation, maintenance, display and manipulation of concept lattices
- Presents application areas such as information retrieval and mining from text and web data as well as rule mining from structured data
- Features two detailed case studies; exploring the content of the ACM Digital Library using an interface that integrates multiple search functionalities; and mining web retrieval results through the system CREDO, a version of which is available on-line for testing
Concept Data Analysis: Theory & Applications is essential for researchers active in information processing and data mining as well as industry practitioners who are interested in creating a commercial product for concept data analysis or developing content management applications. Computer science students will also find it invaluable.
Table of Contents:
I Theory and algorithms.
1 Theoretical foundations.
1.1 Basic notions of orders and lattices.
1.2 Context, concept, and concept lattice.
1.3 Many-valued contexts.
1.4 Bibliographic notes.
2.1 Constructing concept lattices.
2.2 Incremental lattice update.
2.4 Adding knowledge to concept lattices.
2.5 Bibliographic notes.
3 Information retrieval.
3.1 Query modi.cation.
3.2 Document ranking
4 Text mining.
4.1 Mining the content of the ACM Digital Library.
4.2 MiningWeb retrieval results with CREDO.
4.3 Bibliographic notes.
5 Rule mining.
5.2 Functional dependencies.
5.3 Association rules.
5.4 Classification rules.
5.5 Bibliographic notes.