Machine learning is the discipline that trains computers to recognize patterns in data. One application is text classification, which uses computers to recognize patterns in textual data and automatically classifies textual content into categories. In this talk, we will review several health-related case studies using text classification to organize tweets. We will discuss ongoing and emerging health applications in syndromic surveillance, physical activity monitoring, food poisoning, and vaccine sentiment. Through a case study, we will also review the machine learning text classification workflow for this task. We will discuss issues regarding working with tweets, including acquisition, sampling, labeling, encoding, building a classification model, normalizing results, and evaluating classification performance.
nyuinformatics
コメント