dbl-group-3

REQUIREMENT

---

Go to Anaconda Navigator terminal

import regex
import transformers 
import scipy 
import texblob
import wordcloud
import sqlite3

STRUCTURE

---

• main: including building conversation in first sprint

• definitions: python files defining common definitions like airlines, airlines_id, countries code...

• Tools: including all functions we used

  • Cleaning.py: all of the cleaning data functions
  • conversation.py: for building, filtering conversations
  • preprocessing.py: for preprocessing text data
  • sentiment_analysis.py: including functions to run 2 model Roberta and Textblob

• sprint 2: including nb we did in sprint 2
  • Roberta.ipynb: plotting related to sentiment analysis
  • evaluation-Roberta-and-textBlob.ipynb: evaluation of both models
  • Countring conversations.ipynb: counting the conversations
  • DBL sentiment analysis TextBlob: perform TextBlob analysis and textcloud

• demo_sentiment_evolution: sentiment evolution

• demo_sentiment_reply_correlation: final extra 1

• demo_topic_analysis: final extra 2

• geo sentiment: final extra 3

• Cleaning.ipynb: cleaning the data

• conversation.ipynb: extract conversations

• Load the data.ipynb: load, clean and store data

GITHUB

---

• HTTPS: https://github.com/PhatLeBao-Phat/dbl-group-3.git

• SSH: [email protected]:PhatLeBao-Phat/dbl-group-3.git

INSTRUCTION

---

1. Checking the final poster

Run the following notebooks:

• demo_sentiment_evolution/Demo-sentiment-evolution .ipynb
• demo_sentiment_reply_correlation/reply_time_sentimnet_correlation.ipynb
• demo_topic_analysis/demo_topics.ipynb
• geo sentiment/Demo-geo.ipynb
• geo sentimnet/sentiment_location_final.ipynb

CONTACT

---

uni-mail: [email protected]