This document presents an overview of text mining. It discusses how text mining differs from data mining in that it involves natural language processing of unstructured or semi-structured text data rather than structured numeric data. The key steps of text mining include pre-processing text, applying techniques like summarization, classification, clustering and information extraction, and analyzing the results. Some common applications of text mining are market trend analysis and filtering of spam emails. While text mining allows extraction of information from diverse sources, it requires initial learning systems and suitable programs for knowledge discovery.

1. TEXT MINING
Presented By:
Prakhyath Rai
Asst. Professor, Dept. of ISE,
SCEM, Mangaluru

2. Outline
 Introduction
 Data Mining Vs. Text Mining
 Motivation for Text Mining
 I/O Model for Text Mining
 Steps for Text Mining
 Key Terms in Text Mining
 Text Mining Frameworks
 Merits of Text Mining
 Applications of Text Mining
 Demerits of Text Mining
 References
Prakhyath Rai, Asst. Professor, Dept. of ISE, SCEM, Mangaluru-575007

3. Introduction
Text Mining is a Discovery
Text Mining is also referred as Text Data Mining (TDM)
and Knowledge Discovery in Textual Database (KDT).
Text Mining is used to extract relevant information or
knowledge or pattern from different sources that are in
unstructured or semi-structured form.
Prakhyath Rai, Asst. Professor, Dept. of ISE, SCEM, Mangaluru-575007

4. Introduction Cont.
Extract and discover knowledge hidden in text
automatically
Aid domain experts by automatically:
 identifying concepts
extracting facts/relations
discovering implicit links
generating hypotheses
Prakhyath Rai, Asst. Professor, Dept. of ISE, SCEM, Mangaluru-575007

5. Data Mining vs. Text Mining
Data Mining Text Mining
Process directly Linguistic processing or natural
language processing (NLP)
Identify causal relationship Discover heretofore unknown
information
Structured Data Semi-structured & Unstructured
Data (Text)
Structured numeric transaction
data residing in rational data
warehouse
Applications deal with much
more diverse and eclectic
collections of systems and
formats
Prakhyath Rai, Asst. Professor, Dept. of ISE, SCEM, Mangaluru-575007

6. Motivation for Text Mining
Approximately 90% of the world’s data is held in
unstructured formats (source: Oracle Corporation)
Information intensive business processes demand that we
transcend from simple document retrieval to “knowledge”
discovery.
Prakhyath Rai, Asst. Professor, Dept. of ISE, SCEM, Mangaluru-575007

7. Input-Output Model for Text Mining
Input
Text Mining
Technique
Output
Patterns
Connections
Trends
Documents
Prakhyath Rai, Asst. Professor, Dept. of ISE, SCEM, Mangaluru-575007

8. Steps for Text Mining
Pre-Processing the Text
Applying Text Mining Techniques
Summarization
Classification
Clustering
Visualization
Information Extraction
Analyzing the Text
Prakhyath Rai, Asst. Professor, Dept. of ISE, SCEM, Mangaluru-575007

9. Keywords Terms in Text Mining
 Information Extraction (IE)
The science of searching for
Information in documents
Documents themselves
Metadata which describe
documents
Text, sound, images or data,
within database: relational
stand-alone database or
hypertext networked
databases such as the
Internet or intranets.
 Artificial Intelligence (AI)
Artificial intelligence
(AI) is a branch of
computer science and
engineering that deals
with intelligent behavior,
learning, and adaptation
in machines.
Prakhyath Rai, Asst. Professor, Dept. of ISE, SCEM, Mangaluru-575007

10. Merits of Text Mining
Database limits itself to Storage of less Information
whereas Text Mining overcomes this limitation
Extraction of relevant Information and Relationships
from Natural Documents
Extraction of Information from Unstructured or Semi-
structured Documents
Prakhyath Rai, Asst. Professor, Dept. of ISE, SCEM, Mangaluru-575007

11. Applications of Text Mining
Analysis of Market Trends
Classification Technique
Information Extraction Technique
Analysis and Screening of Junk Emails
 Classification on the basis of pre-defined frequently
occurring items
Prakhyath Rai, Asst. Professor, Dept. of ISE, SCEM, Mangaluru-575007

12. Demerits of Text Mining
Requires Initial Learned Information System for
Initial Extraction
Suitable programs are not been defined to Analyze
Text from Mining Knowledge or Information
Prakhyath Rai, Asst. Professor, Dept. of ISE, SCEM, Mangaluru-575007

13. References
[1] R Baeza-Yates and B Ribeiro-Neto. “Modern Information Retrieval”, ACM
Press, New York, 1999.
[2] Ning Zhong, Yuefeng Li and T. Grance, “Effective Pattern Discovery for Text
Mining,” IEEE Transactions on Knowledge and Data Engineering, Vol. 24, No. 1,
January 2012.
[3] Raymond J Mooney and Un Yong Nahm, “ Text Mining with Information
Extraction”, Proceedings of the 4th International MIDP Colloquium, pages 141-
160, Van Schaik Pub., South Africa, 2005.
[4] M E Califf and R J Mooney, “Relational Learning of Pattern-Match Rules for
Information Extraction”, Proceedings of the 16th National Conference on Artificial
Intelligence (AAAI-99), pages 328-334, Orlando, FL, July 1999.
[5] D Freitag and N Kushmerick, “Boosted Wrapper Induction”, Proceedings of
the 17th National Conference on Artificial Intelligence (AAAI-2000), pages 577-
583, Austin, TX, July 2000.
Prakhyath Rai, Asst. Professor, Dept. of ISE, SCEM, Mangaluru-575007