IDA-Caffeine-Overflow / IDA_NLP Public

Notifications You must be signed in to change notification settings
Fork 2
Star 0

Audio is sent to DEEP-LEARNING 7-LAYERED DNN Model , which considers spectrogram of the audio and uses the Librosa library for classifying the particular emotions Angry, Sad, Disgust, Surprised, and a testing accuracy of 75%.

0 stars 2 forks Branches Tags Activity

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 7 Commits
PriorityText.ipynb		PriorityText.ipynb
PriorityText_Alternative.ipynb		PriorityText_Alternative.ipynb
TweetSummary.ipynb		TweetSummary.ipynb
readme.md		readme.md

Repository files navigation

Natural Language Processing for Text Analysis of Call Transcript and Tweet.

The text is prioritized using TF-IDF, Page Ranking, Cosine Similarity using steps:

Pre-Processing-> TF-IDF Matrix -> Similarity Semantics -> Page ranking.

The tweets are prioritized using Integer Linear Programming methods like Primal and Dual, and using library such as

Textacy, Spacy, NLTK, PyMathProg.

Tweet Summarization is Achieved in 3 steps:

1.I want the total length of all the selected tweets to be less than some value L

2.If I pick some content word (out of my possible content words) , then I want to have at least

one tweet from the set of tweets which contain that content word, .

3.If I pick some tweet i (out of my possible tweets) , then all the content words in that tweet are also selected.

About

Audio is sent to DEEP-LEARNING 7-LAYERED DNN Model , which considers spectrogram of the audio and uses the Librosa library for classifying the particular emotions Angry, Sad, Disgust, Surprised, and a testing accuracy of 75%.

nlp spacy nlp-machine-learning textacy nltk-python

Custom properties

Report repository

Releases

No releases published

Packages

No packages published

Languages

Jupyter Notebook 100.0%