Inspiration
Feedback is valuable information to take decision. For top companies ‘continuous improvement’ is not just a showy catchphrase. It’s a true focus based on feedback from across the entire organization. Our Inspiration is developing a feedback collecting and reviewing application for any company of users wish with high percentage of accuracy in classifying the reviews.
What it does
Feedback AI gets company name, feedback and rating from the employees of any company and store it in database. When user ask for review about a company, all the feedback got for that company will be taken and passed one by one to emotional trait and the emotion found would be passed to sentiment analysis. Based on overall sentiment score the feedback classification will be displayed if it is not a hate speech. User can also download the review result of feedback of the company.
How we built it
We have collected the company name and feedback from the employees and stored it in database using sqlite3 library in python. The feedback will be fetched from the database for the selected company. It will be passed to expert.ai Hate Speech detection to find out hate speech and avoid it from displaying. The feedback will then be passed to expert.ai Emotional trait to classify the category of emotion that the feedback posses. The result of the emotional trait will be given to expert.ai Sentimental analysis. Based on the overall sentiment score the feedback will be put into corresponding sentiment(positive, negative, neutral) in a python dictionary.
Challenges we ran into
Handling null values while entering values to SQL command. No output category returned by Emotional trait as the sentence does not contains it.
Accomplishments that we're proud of
We are proud of developing fully functional application useful of organizations as well as employees and freshers to know about the company.
What we learned
Developing and deploying web application with streamlit. Integrating API to application. Working with sqlite3
What's next for Feedback AI
Including location details of the company and adding more filters for searching reviews.
Log in or sign up for Devpost to join the conversation.