A scientist named Suresh Venkakatasubramanian from University of Utah’s school of computing along with other fellow scientists are working to find out if a machine algorithm is biased or not. How can a machine algorithm or a computer program be biased? Nowadays, almost every recruitment, job hiring, application selection process is done online. It has been known that mist of the application processes use algorithm that automatically select candidate profiles based on some keywords of skills, age, grade or marks etc. These criteria are mandatory while selecting a candidate for recruitment. But it has been seen that many computer program are written in a biased way which further discriminates candidates based on gender and race. Selection based on race and gender is highly unethical and hence research is being done to point out if an algorithm has such loopholes.
As mentioned earlier, almost everywhere, online applications are used and hence recruitment algorithms are very much in use these days. So it is of prime importance. Those computer algorithms and computer programs are written in such a way that they automatically sort out applications based on their keywords if skills and eligibility. Venkatasubramanian said that their main aim of the research is to dig deep into an algorithm and find out if the particular computer program is violating the legal standards for fair access to resources by people from all race and gender.
“There’s a growing industry around doing résumé filtering and résumé scanning to look for job applicants, so there is definitely interest in this,” says Venkatasubramanian. “If there are structural aspects of the testing process that would discriminate against one community just because of the nature of that community, that is unfair.”
Source: University of Utah