Top tech companies have yet to close the gender gap in hiring, a disparity most pronounced among technical staff such as software developers where men far outnumber women and other under-represented genders.
What it does
Enable unbiased screening of resume by hiding sensitive information like name gender, email which describes the ethnicity, religion or gender of a person until the shortlisting so that everybody gets equal chance based entirely on skills. To give an inclusive idea about the organization's opportunities we help check job descriptions and advertisements for dominating gender-specific words and review along with suggesting user appropriate replacements
How we built it
We built a website using react and used the spacy NER(name entity recognition) algorithm to run in the backend and produced the desire results and classify the resume data into entities like name, email, skills etc. Further, we hid the data of consideration and produce a text file of other important things. We have deployed the model using Heroku.
Challenges we ran into
Applying the algorithm to data was a tough challenge. We were unable to implement our second feature of real-time job adds review.
Accomplishments that we're proud of
We are happy that we could at least implement one feature and make a clean working UI.
What we learned
We learned something new at every step like working with spacy, dockerizing and deploying. The NER algorithm was pretty challenging and useful.
What's next for IncluGen
We aim to implement the leftover feature and also make a chrome extension of it. We plan to include more features and optimise our algorithm for better accuracy