What steps should I take to create a sentiment analysis web application using Twitter?
-
I am planning to create a sentiment analysis engine but after searching lots of blogs/technical papers/articles I am still confused about how to create this project. The following are my concerns: Data mining problem: found out that data (tweets) for analyzing sentiments is provided by Twitter's API (Twitter4J). I'm not sure how to use this API along with other parts of my project (I am planning to use Stanford CoreNLP or Python NLTK). Secondly, I have to use a lexicon resource like SentiWordNet or other resource available for analyzing the Tweets by assigning polarity to words. How do I use this in conjunction with Twitter API & Stanford CoreNLP/Python NLTK). Also, I have to apply Naive Bayes Algorithm for polarity prediction. If I use Python NLTK, how do I provide the user interface to it in context of a web application? For example, if I use Stanford CoreNLP, can I code the sentiment analysis prediction part in Model part of MVC since SCL is Java based and code for View(GUI) part in JSPs using JSP and Servlet technology considering MVC pattern? So basically, how do I implement GUI for providing an input text(query term to fetch tweets related to input text and providing analysis of fetched tweet by indicating polarity) if i use Python? Is there any database required to build the project? Which would you recommend: Stanford CoreNLP or Python NLTK?
-
Answer:
Please can i use NLTK for sentiment analysis ... You must be signed in to read this answer.Connected to GoogleConnected to FacebookBy continuing you indicate that you have read and agree to the . Loading account...Complete Your ProfileFull NameChecking...EmailChecking...PasswordChecking...By creating an account you indicate that you have read and agree to the .
Silhadi Ismahane at Quora Visit the source
Other answers
This is not an expert answer but from a grad student 1. $ easy_install twitter import twitter twitter_search = twitter.Twitter(domain="http://search.twitter.com") Then you can explore i guess.. 2. They are 2 approaches for sentiment analysis. (a) Machine learning based: Use it when you have a domain specific problem. NLTK has many examples and functions for learning based analysis. (b) Lexicon based: Use it when you want a generic solution but ready for more error. For lexicon approach use NLTK combined with sentiwordnet. The steps are: tokenize words, lematize them (gud -> good), POS tag, pick the adjectives and assign rating to them. Take care of negators (not good) and intensifiers (very good). The rating may be binary (positve, negative) or real value (stars). With this you can expect a reasonable accuracy. 3. You can plug the developed python+NLTK code into Django project. Use HTML5 for GUI and use python to process the request 4. For training, yes. For the final implementation, no. 5. Python + NLTK
Rajalingappaa Shanmugamani
Related Q & A:
- What steps do I take so I'll be able to travel to other countries in Europe?Best solution by Yahoo! Answers
- What steps does it take to be a police officer?Best solution by Yahoo! Answers
- What classes should I take to become a social worker?Best solution by Yahoo! Answers
- What steps should I take to become an Obstetrician gynecologist?Best solution by Yahoo! Answers
- What steps do i take to become a Registered Nurse?Best solution by Yahoo! Answers
Just Added Q & A:
- How many active mobile subscribers are there in China?Best solution by Quora
- How to find the right vacation?Best solution by bookit.com
- How To Make Your Own Primer?Best solution by thekrazycouponlady.com
- How do you get the domain & range?Best solution by ChaCha
- How do you open pop up blockers?Best solution by Yahoo! Answers
For every problem there is a solution! Proved by Solucija.
-
Got an issue and looking for advice?
-
Ask Solucija to search every corner of the Web for help.
-
Get workable solutions and helpful tips in a moment.
Just ask Solucija about an issue you face and immediately get a list of ready solutions, answers and tips from other Internet users. We always provide the most suitable and complete answer to your question at the top, along with a few good alternatives below.