Artificial intelligence for rule-based filtering with Python

I have an academic project to create an application to define a consumer's "feel" for brands, be it positive, negative or neutral. However, I am stuck with no idea how to get a sample on how to "read" English with code. For example:

Ms. Darley buy me an apple :)

Contrast to the next sentence:

I loved Darley so much and he woke me up every morning.

From human common sense, we can see that the first sentence is about "man." The first person to describe a lady named Mrs Darley, but in my case I only want to focus on getting only brand names, which means the first sentence should be dropped, mentioning the second sentence instead. From this sentence, we can determine that it is the object that most likely describes the toothpaste we use every early morning.

How can I build this "common sense" with Python? (If there is any example in PHP I could try converting it to Python) or is there a library available? Alternatively, please suggest some keywords to research.

+3


source to share


1 answer


Thanks to Tom De Smedt and Walter Daelemans of the CLiPS Computational Linguistics Group at the University of Antwerp. They resolved my question in this Journal with an example and full source code on this website page .

The concept of this application is the first application associated with crawling from a social site or any publicly accessible websites, and then a backend connected to the English Wordnet.org database to determine the "type" or "category" of words from sentences. Then follow this classification method. it is amazing!!



Thanks for the answer. and hope this helps others :)

+2


source







All Articles