UCU Summer School NLP projects

•••► Topic: UCU Summer School NLP projects •••► Speakers: Vitalii Radchenko and his UCU teams •••► Who will be interested: everyone familiar with the basics of Machine Learning and NLP •••► Where: Grammarly Office, 13 Tereshchenkivska St., Apt 110, 8th floor •••► When: Wednesday, August 9th, 7 PM •••► Language: Ukrainian, Russian •••► Registration: https://gram.ly/DNZR During the summer school, UCU students did some cool projects and would like to present them to the Kyiv machine learning and NLP community to share their interesting results. •••► First Project: Transfer Learning for Sentiment Analysis To create a good model, we need a lot of data. But what if we don't have it? For example, what if a company wants to determine its customers' attitude, but we don't have enough data to train a good model on a Ukrainian or a German corpus? It is possible to do so with transfer learning from English to German and from Russian to Ukrainian as we usually do it in computer vision. •••► Second Project: Data Science Trends and message classification on OpenDataScience Slack data OpenDataScience is the biggest data science community, and it is interesting to explore Data Science trends with SNA based on channels activity. Moreover, we will show how to predict question response time, how to find the most useful people in a community and how to classify messages. It is a multi-class classification of 12 channels and 100 users with different NLP approaches.