Sentiment Analysis of Customer Reviews of various products on e-commerce websites The main purpose of the service is to analyze the sentiment's (good,bad,disgusted,angered,etc.) from the textual reviews available on popular e-commerce websites and to classify a product if it is a 'Success' or 'Failure', based on certain pre-decided factors. The service also provides mechanism to translate a given review supposing in other language to be translated into English (the default language used by IBM Blumix Tone Analyzer) and there onwards carry the sentiment analysis process.
Tech Stack :
- Languages used - Python
- API's used - IBM Bluemix Tone Analyzer , Messenger , Slack, IBM Bluemix Language Translator.
Mainly two approaches will be discussed over here - a) Using threshold valued-classifier. b) Using IFTTT.