CODEWITHSUNDEEP
data-miningtext-mining
Upendra Siripurapu
Upendra Siripurapu

Reputation: 316

need an idea about text mining for mining data from bulk of files

I am new for data mining. I am doing my B.Tech final year, my final year project title is "Extraction and analysis of faculty performance of management discipline from student feedback using text mining". Here we will have number of files which contains feedback given by students, each student will have one single file. From all these files we have to retrieve useful information.

can any one suggest me how to start, what are the tools to be used? what are the technologies to be used? I am familiar with JAVA(jse), can i achieve this using java programming language, how?

regards... Upendra.S

Upvotes: 1

Views: 1472

Answers (3)

Debaditya
Debaditya

Reputation: 2497

In my opinion , you can go through these topics. It helped in one of my projects.

Tool - Matlab TMG tool .

Algorithm Used - Latent Semantic Analysis . (LSA Example)

Mathematical Concept - Singular Value Decomposition . (SVD)

Upvotes: 0

Neil McGuigan
Neil McGuigan

Reputation: 48287

Some ideas:

  • which are the most frequently used words or phrases?
  • which words often co-occur (association analysis)
  • word/phrase frequency by student grade (which words do the top and bottom students use in their prof reviews?)
  • word/phrase frequency by faculty rating (which words are associated with the score that students give the faculty member?)
  • word/phrase frequency by faculty tenure (which words are associated with new and older profs?)

here is my five part series on text mining with rapidminer:

http://vancouverdata.blogspot.com/2010/11/text-analytics-with-rapidminer-loading.html

Upvotes: 3

Has QUIT--Anony-Mousse
Has QUIT--Anony-Mousse

Reputation: 77495

I believe Rapidminer has a text mining extension. Then there is JTMT.

Also use the search function here:

Also have a look at the Educational data mining community. They probably have similar tasks.

Upvotes: 1

Related Questions