digital-cognition-co-in / DigitalCognition

DigitalCognition - Main repo - Core Product
2 stars 2 forks source link

Twitter , Tweepy , Social Network Analysis #35

Open RohitDhankar opened 4 years ago

RohitDhankar commented 4 years ago

1/ Gathering tweets from Twitter , using Tweepy 2/ Creating a Social Network , mining only this Social Network for TimeSeries twitter data 3/ Analysis of tweets - extracting signal for Corporate Performance

@noob-master147 , @subSri and @rish07 ( Rishi Raj Singh ) , to collab on all three sub tasks. Reach out to @RohitDhankar , for clarifications . Some old code shared by Rohit with @subSri , earlier.

@noob-master147 - to see this video for background on tasks -- https://www.youtube.com/watch?v=Jo2rZZqdHZ4

RohitDhankar commented 4 years ago

Am adding some reads and old code files here - these were shared with @subSri earlier - 1/ WORD FREQUENCY AND SENTIMENT ANALYSIS OF TWITTERMESSAGES -- This can be used as a refrence for Sentiment Analysis of Tweets Collected after we have done the SNA – Social Network Analysis of the Tweets ... https://arxiv.org/pdf/2004.03925.pdf

2/ Research on sentiment analysis for Twitter textual data,which tackles the problem of analyzing the messages posted on Twitter in terms of the sentiments they express, has alsobeen performed. Twitter is a very challenging domain for sentiment analysis mainly due to the length limitation of texts[12]. The majority of past approaches employed a traditional machine learning methods such as logistic regression, SVM,MLP, etc., trained on lexicon features and sentiment-specificword embeddings (vector representations of words) [13], [14].More recent approaches typically opted for sequence learning models trained to learn relevant embeddings for classification from large pretrained word embeddings, particularly the GloVe[15] embeddings for Twitter data. Best performing models of this breed include Cliche (2017) [16] and Baziotis et al.(2017) [17], which shared the first place for sentiment analysis on Twitter (Task 4A [18]) at the International Workshop on Semantic Evaluation 2017 (SemEval-2017). The noveltyin our approach to Twitter sentiment analysis involves the implementation of state-of-the-art transformer methods such asBERT [19] and RoBERTa [20], whose outstanding sentiment classification prowess remains untested for Twitter data in theliterature https://arxiv.org/pdf/2004.03688.pdf

3/ https://paperswithcode.com/paper/stock-price-correlation-coefficient

4/ https://link.springer.com/content/pdf/10.1007%2F978-3-319-94463-0.pdf

5/ 2 PDF Books shared with @noob-master147 and @subSri

6/ Social Media Analytics / Social Network Analysis --https://rstudio-pubs-static.s3.amazonaws.com/107167_28714fdd899a475cb6d07c1c31502a10.html https://github.com/RohitDhankar/Social-Network-Analysis_Python-and-RStats/blob/master/TwitterSNA.R https://datasciencewithrandpython.blogspot.com/search/label/Twitter%20API

https://towardsdatascience.com/generating-twitter-ego-networks-detecting-ego-communities-93897883d255 https://www.datacamp.com/community/tutorials/social-network-analysis-python https://medium.com/future-vision/visualizing-twitter-interactions-with-networkx-a391da239af5 https://nycdatascience.com/blog/student-works/find-twitter-influencers-through-social-network-analysis/ https://blog.f-secure.com/why-social-network-analysis-is-important/ https://www.kdnuggets.com/2016/06/mining-twitter-data-python-part-1.html https://github.com/topics/social-network-analysis?l=python&o=asc&s=stars https://ai-journey.com/2019/11/who-are-the-top-hr-analytics-influencers-on-twitter/

RohitDhankar commented 4 years ago

@subSri == https://rawgit.com/wesslen/summer2017-socialmedia/master/day2/twitter-social-networks.html