Our project for Smart India Hackathon 2018 contributed by a team of 6 geeks.
please keep adding your views in this file
Taks to-do:
- Get api's of reddit, twitter, quora to fetch tweets/content to be analysed in the later part
2. Gather a training set with tweets/content with their sentiments analysed to be used as a training set
3. Use word2vec and other NLP tools to gather insights from the fetched content
4. Apply supervised learning models to train our model
Potential solutions for sentimental analysis that have excellent, and proven results
https://github.com/cbaziotis/datastories-semeval2017-task4 https://github.com/bfelbo/DeepMoji
Working
- We input modi in our system
2.It gathers material from the 3 social netwroking handles and using word2ves gathers the top 3 related content related to it.
Let's say the three keyqords came as : PM Modi, Nirav Modi, Modi sarkar
3. Now, user selects the desired keyword and then our system fetches a large amount of tweets/content and analyses sentiments in it and shows the insights of the results generated.
Ab kaam start karo nhi toh train ticket cancel karo :P