#Project-C: It is a group project of 5 members and I've worked on Randomn forest & SVM. Here, we present a computational framework for automatically quantifying verbal and nonverbal behaviours in the context of job interviews. The proposed framework is trained by analysing the videos of 138 interview sessions with 69 internship-seeking undergraduates at the Massachusetts Institute of Technology (MIT). Our automated analysis includes language (e.g., word counts, topic modelling), and prosodic information (e.g., pitch, intonation, and pauses) of the interviewees. Our framework can automatically predict several other high-level personality traits such as engagement, friendliness, and excitement and can quantify the relative importance of prosody, language information. Implementation of various machine learning algorithms like k-fold cross validation and regression methods like Lasso, Random Forest and SVM for feature extraction and classification on the dataset has been tried to improve the accuracy and to predict interview ratings and the likelihood of hiring using extracted features. Further goal is to work on optimization algorithms considering the accuracy factor.