Antimony

Main Banner
The Applicant answering questions in the applicant sided program
Code Generated by Application sided program
Now, the Recruiter inputs the code in his software, which puts the answers in the Naive Bayes model, which gives a score out of 10.

Who we are

Hi! I'm Aryan! I'm a 17 year old student studying in 12th Standard in Singapore, majoring in Computer Science. Sai is a 16 year old student also studying in Singapore majoring in Biology. We both have one passion: coding, which is what brought us to AlphaHacks 2021.

The idea

It's a known fact that employers are now looking for more than just skills. They are looking for moral values like kindness, compassion, empathy from their employees. But, these value based questions can be just as well collected and processed with an AI model, and are ignored by current AI recruitment platforms. So, here's Antimony, a value based recruiting software powered by the Naive Bayes Classification system. This software can save recruiters precious time by automating the value based recruiting process.

What it does

Antimony has 2 components: one program for the applicant, and one program for the recruiter. The applicant sided program asks these 'value-based' questions to the applicant. After all the questions have been answered, The program generates a 'code', which the Applicant is supposed to provide to the employer.

Now, upon receiving the code, the recruiter puts the code in the recruiter's program, which then analyses the answers and predicts if the employee has good moral values. The applicant's moral based values are rated from a score of 0-10, with 0 being the lowest value and 10 being the highest value.

How we built it

I used python with a number of data science modules included numpy, sklearn and pandas.

Challenges we ran into

Just a few hours ago, one of my datasets got corrupted due to a file error. I had to start from scratch which was very time consuming.

What does it bring on the table?

It brings a whole new perspective to AI-based recruitment, It asks you value based questions which are often ignored in traditional AI recruitment softwares. This is not a substitute to AI based recruitment, but rather an addition to AI based recruitment and potentially, physical recruitment.

Accomplishments that we are proud of

We made a working code in a relatively short amount of time.

What we learned

Time management is very important for a successful project.

What's next for Antimony

I wanted to implement Natural Language Processing in Antimony, but due to time constraints I wasn't able to do so. NLP will be implemented in the future. Another improvement will be the frontend. I will push the frontend to Django, a web based python frontend which can be accessed by everyone. Sai wants to implement Deep Learning algorithms in Antimony to make the predictions more accurate.