I have a requirement for which I do not know if NL can do it. Plz advise.
My requirement is to scan a sentence in english language and figure out some qualitative info about it. Such as , what are the subjects , nouns in the sentence and what is said about them is a descriptive , suggestive or does it affects positively or negatively.
As an example, lets say I have a Fan page in facebook , and someone posts a post on my wall. I need to know if the post says something good or bad about me and accordingly I can map it to a perception scale from , say -10 to +10.
Can something like this be done with with Natural language processing toolkits?
If no, then what is the way.
Thanks in advance
Shyam
Yes, Natural Language Processing methods can do what you ask for:
To find out which are the “nouns” in a text is called POS-Tagging.
Identifying the syntactical strucure of sentences is called parsing. Depending on your parsing method and how strict the syntax of the language is you might want to look into semantic role labeling aswell to find subjects/objects in a sentence. Classifying the polarity (positive/negative) of a statement is called sentiment analysis.
The Python NLTK provides some tools you can get started with, but sentiment analysis is an active research area and a task where a lot of other NLP-methods have to work together, so it is certainly not the easiest field to get started with in NLP. Anyways, a survey of the academic research in the field can be found in Pang & Lee (2008).