Text modules

Researching and verifying our project

This week I researched on ways to make the project come to light. Firstly I had to find some libraries to do text analysis. Some modules that I found were, spacy and NLTK. The libraries allows us to clean, tokenize and lemmitize our text fairly easily once you have gotten the theory parts of the concept right. This allows the computer to “read” the information that it has been given as all of the text has been cleaned, transformed to their base words and finally converted to vectors/numbers that the machine can comprehend.

We also wanted to do a feature where if a user wanted to know more about a particular aspect of the product. The user could just type that into the textbox, and the machine will try to write about that aspect of the product. This requires the use of an RNN or CNN, but I am not so sure as I am fairly new to neural networks. More research is needed to be done.

End

That is all for this week. See you next Week!