Skip to content

Natural language processing

Islam Sediek Tolan edited this page Mar 23, 2015 · 10 revisions

Summary

Using (tm Package and NLP package) to answer user's questions.

Description

We will read a question from a web interface using (tm Package) then use (NLP Package) to break the question into tokenizers and convert it to matrix. then we can train a model to predict score using machine learning algorithm . finally use a UIMA to discover Knowledge to understand the question and search a database for the answer. This system can be used to build intelligent expert system in various fields.

Related work

This system can be a good beginning to more intelligent expert systems.

Potential tasks

1- Create a web interface to enter the questions.

2- Use (tm package)to read the question.

3- Use (NLP package) to convert the question to tokenizers then the matrix finally train
model and predict score.

4- Use UIMA to discover Knowledge.

5- Create database that will save answers for the questions.

Skills required

a very good programming experience in R language and Java.

Good programming Knowledge in HTML.

Database and SQL.

Knowledge about tm package and NLP package.

Test

Assume that you have a text (try any text) by using tm package load the text for processing after that use (Bag of word model) to represent this text

Mentor

Name : Islam sediek Email : islamsedtolan1@gmail.com