DEVELOPMENT OF AN AI-BASED INTERVIEW SYSTEM FOR REMOTE HIRING

IEEE Transactions on Affective Computing, 2016

We present a computational framework for automatically quantifying verbal and nonverbal behaviors in the context of job interviews. The proposed framework is trained by analyzing the videos of 138 interview sessions with 69 internship-seeking undergraduates at the Massachusetts Institute of Technology (MIT). Our automated analysis includes facial expressions (e.g., smiles, head gestures, facial tracking points), language (e.g., word counts, topic modeling), and prosodic information (e.g., pitch, intonation, and pauses) of the interviewees. The ground truth labels are derived by taking a weighted average over the ratings of 9 independent judges. Our framework can automatically predict the ratings for interview traits such as excitement, friendliness, and engagement with correlation coefficients of 0.75 or higher, and can quantify the relative importance of prosody, language, and facial expressions. By analyzing the relative feature weights learned by the regression models, our framework recommends to speak more fluently, use less filler words, speak as "we" (vs. "I"), use more unique words, and smile more. We also find that the students who were rated highly while answering the first interview question were also rated highly overall (i.e., first impression matters). Finally, our MIT Interview dataset will be made available to other researchers to further validate and expand our findings.

Journal of Management Studies and Development

This research paper examines the applicability of artificial intelligence (AI) in the recruitment and selection process, with a specific focus on the Interview phase. The study conducted qualitative interviews with 17 recruitment professionals, including hiring managers, recruiters, and HR executives, to gather their perspectives on AI adoption. The Unified Theory of Acceptance and Use of Technology (UTAUT) framework was utilized as the theoretical model to guide the research. The findings indicate that the recruitment professionals were generally receptive to using AI in various recruitment phases. However, when it came to the interview phase, there was a distinct lack of enthusiasm for AI adoption among the professionals. They expressed concerns about providing a better candidate experience and perceived AI-based interviews as a lack of commitment to candidates. As a result, regardless of the hiring volume or job group, the professionals were hesitant to integrate AI into the inte...

Proceedings of the 28th International Conference on Intelligent User Interfaces

The most popular way to assess talent around the world is through interviews. Interviewers contribute substantially to candidate experience in many organizations' hiring strategies. There is a lack of comprehensive understanding of what makes for a good interview experience and how interviewers can conduct candidate-centric interviews. An exploratory study with 123 candidates revealed critical metrics about interviewer behavior that affects candidate experience. These metrics informed the design of our AI-driven SmartView system that provides automated post-interview feedback to Interviewers. Real-world deployment of the system was conducted for three weeks with 35 interviewers. According to our study, most interviewers found that SmartView insights helped identify areas for improvement and could assist them in improving their interviewing skills.

Springer, 2021

The interview and hiring process for an average IT applicant clocks in at about 33 days. Currently using the outdated man-driven SOP, companies, and candidates lose out on precious time and resources which could be utilized for critical operations and tasks. A contemporary application for the automation of base-level interviews has been developed using open-source machine learning (ML) libraries that intensively investigate the candidate’s speaking patterns and facial orientation to generalize a report of the candidate. A general score and rating have been generated that accurately reflects the level of proficiency of the interviewee in terms of soft skills and body language. This application is named HRPro. The application can successfully conduct a short 5 min interview and generate a post-processing report. The report is ideally custom mailed manually to the applicant at the end of the processing time. Information consists of emotion analysis, interview points over time graph, and facial orientation over time graph. Interview points are assigned against a log of 100 separate good and objectively bad interviews used to train the model using pseudo ML algorithms.

IAEME Publications, 2021

As AI (artificial intelligence) technology advances, global enterprises pay keen attention to AI-based interviews as a new employment method. Mainly, untact AI interviews are widely spread around the globe in this era of COVID-19 pandemic. Accordingly, this study examines factors that af ect AI technology application to corporate job interviews and reason out key factors. To this end, this study proposes a decision-making model that can be a vital standard for AI technology application to corporate job interviews. For decision-making modeling, the Analytic Hierarchy Process (AHP) method was used. Eighteen major factors were derived from previous studies based on the unified theory of acceptance and use of technology in 4 areas: 'performance expectancy,' 'effort expectancy, 'social influence,' and 'facilitating conditions.' A survey was conducted by total 40 professionals as user and provider related AI job interview solutions. The survey results showed that the most significant effect of adopting an AI-based job interview system included ease of use, job fitness, perceived usefulness, and perceived compatibility. Therefore, more attention should be paid to introducing an AI job interview system efficiently as an internal recruiting system of an organization rather than external environments or conditions

Business Communication Research and Practice

2020

The introduction of artificial intelligence has reshaped job application and selection practices. The use of the internet in the organization has efficiently revolutionized the way of recruitment in an efficient manner. We aim to optimize the recruiting process with AI. One-way interviewing is now a standard method for screening in the first round providing an efficient, fair, and structured method for conducting interviews. E-interviews can be carried out in a personal location such as a home anytime. When it comes to hiring, it does not cut jobs but accelerates the process by removing manual tasks and make recruiters far more productive. The field of computer science AI can be used to quickly fill positions but with the right person. The bot can automate the process of hiring from screening to interview scheduling. This can benefit the company with cost-reduction, time-efficiency, accuracy and consistency during the recruitment. The system is smart enough to capture facial express...

arXiv (Cornell University), 2023

Many companies and organizations have started to use some form of AI-enabled automated tools to assist in their hiring process, e.g. screening resumes, interviewing candidates, performance evaluation. While those AI tools have greatly improved human resource operations efficiency and provided conveniences to job seekers as well, there are increasing concerns on unfair treatment to candidates, caused by underlying bias in AI systems. Laws around equal opportunity and fairness, like GDPR, CCPA, are introduced or under development, in attempt to regulate AI. However, it is difficult to implement AI regulations in practice, as technologies are constantly advancing and the risk pertinent to their applications can fail to be recognized. This study examined deep learning methods, a recent technology breakthrough, with focus on their application to automated resume screening. One impressive performance of deep learning methods is the representation of individual words as lowdimensional numerical vectors, called word embedding, which are learned from aggregated global word-word co-occurrence statistics from a corpus, like Wikipedia or Google news. The resulting word representations possess interesting linear substructures of the word vector space and have been widely used in downstream tasks, like resume screening. However, word embedding inherits and reinforces the stereotyping from the training corpus, as deep learning models essentially learn a probability distribution of words and their relations from history data. Our study finds out that if we rely on such deep-learning-powered automated resume screening tools, it may lead to decisions favoring or disfavoring certain demographic groups and raise ethical, even legal, concerns. To address the issue, we developed bias mitigation method. Extensive experiments on real candidate resumes are conducted to validate our study.

Journal of emerging technologies and innovative research, 2019

This paper presents a review on various techniques on predicting the interview performance autonomously using machine learning techniques and analyze the audio-visual features using an interface-based video interview by a virtual agent. Different scenarios are studied and compared for a better understanding of human behavior in case of the video interview. In this paper, the existing optimization techniques are reviewed according to their methodologies and advantages. Based on the review, research gaps are observed and the references are provided for further research work. Index Terms Interface based video interviews, Automatic job prediction, Machine learning for job prediction, Nonverbal behavior prediction, Regression, Deep learning for audio-visual predictions.