Human Sales Ability Estimation Based on Interview Video Analysis

Abstract

In this paper, we introduce our proposed method for estimating the potential sales ability of a person based on interview analysis. A deep learning model was built and trained to estimate the major personality traits according to the OCEAN model. The model utilizes CNNs and LSTMs to extract and learn features from the sequence of the frames for each video. Base knowledge has been built to define the decision rules for classifying the participants depending on their personality analysis and to give the final classification score. The decision rules depend on the relation between the Big Five Factors that describe salesmen represented by high levels of (Extraversion, Conscientiousness, and Openness) and low levels of (Neuroticism and Agreeableness). We collected our dataset by building our website to let participants record self-introduction videos while answering customized questions for our task. The model was trained on two datasets: the CVPR First Impressions V2 was used to train the model to be able to estimate personality traits and our dataset that was used to fine-tune the model to our task and capture the potentials of a salesman. The proposed method has been tested on a variety of samples including video samples for real sales managers and achieved good results in classifying them.

Keywords

• ocean model neural networks computer vision human analysis