You are on page 1of 1

1) Detecting user emotions from Emails or Document files.

- Data preprocessing
- tokenizers
- the algorithm trains the linear classifier on transformed labeled data. It uses
Linear SVM with squared hinge loss for classification. The results show that the
protocol outperformed other existing techniques in domain adaptation analysis.
2) voice to multi class emotion analyzer using Neural network.
- HMM technique
3) Building Knowledge based recommendation engine for OTT platform.
-Netflix Dataset
-Knowledge-based Recommendation Engine comes in handy to solve the cold-start
problem and offer a reliable solution for the business to grow. However, it comes
up with its own form of challenges.
-impact of genre granularity
-Improvising the system:
During navigation, users provide feedback which are termed as “critiques”.
Critiques are then used to help users navigate among the retrieved relevant search
results. There are three forms of critiques:
Unit Critique, represents a change request on a single item attribute.
Compound Critique, represents more than one change request at a time.
Dynamic Critique, takes into account history of user’s critique.
Based on these critiques, the parameter of the query is changed or modified to help
the user meet their choice of results.
-MAE and precision for performance evaluation.
4) Building Sentiment Analyzer for Different categories Brands from Social Network.

You might also like