A tool to analyze feedbacks
Explore the docs »
View Demo
Feedback for any organisation is very important. Feedback helps them to grow, to be better at the services that they provide. As every event demands feedback from the participants or the users, so processing of all the feedbacks by a human is getting difficult. Because of human error, we might as well miss out on the important feedbacks. To reduce human intervention, avoid human error while processing, Feedback Analyzer helps as a tool which will generate an output which will segregate the feedbacks as positive, negative, or neutral and also count the frequency of most repeated words.
Feedback Analyzer uses PySpark for counting frequency of words and a combination of libraries - textblob, vaderSentiment for sentiment analysis. It also generates a wordcloud image for showing most popular words.
- Apache Spark : PySpark
- Sentiment Analysis : textblob, vaderSentiment
- WordCloud : wordcloud
Get all Setup Resources : Feedback_Analyzer_Resources
- Requirements include: Anaconda Navigator(Python 3), JDK, Apache Spark, winutils
- Detailed Steps of Setup: Apache_Spark_Setup
- Requirements include: Anaconda Navigator(Python 3), Editor, Gitbash
- Detailed Steps of Project Setup: Feedback_Analyzer_Setup
- List of packages for Python Environment : numpy, pandas, matplotlib, textblob, vaderSentiment, wordcloud
- Input CSV file with feedbacks having column name 'Feedback'.
- Opening terminal with environment, enter instruction:
spark-submit feedback_analyzer.py sample.csv
- Output generated in the form of zip file
- Aditya Kotkar - ADI-KOTKAR | Email
- Project link: Feedback_Analyzer
- License