Skip to main content

This is a Light weight Open source Python Package made specially for Job Seekers and Recruiters in the field of Data science and Machine Learning (Currently), In order to use the Power of Machine Learning to help Shortlist the Right Candidates for a specific Job.

Project description

PyRecruiter

logo

What is PyRecruiter?

This is a Light weight Open source Python Package made specially for Job Seekers and Recruiters in the field of Data science and Machine Learning (Currently), In order to use the Power of Machine Learning to help Shortlist the Right Candidates for a specific Job.

It works according to the right Hiring standards and helps Recruiters to automate the task of going through Hundreds of application and in addition it has it's own Algorithm which focuses on **Parameters like:

  • Required Skills/Libraries/Frameworks
  • Application based projects on the Required Skills/Libraries/Frameworks
  • Domain Centric Projects
  • Story-telling Projects
  • Good Certifications(Less Weightage)
  • Candidates Looking for Domain change towards ML/AI

These parameters play an important role in searching for the right Data Scientist or a Machine learning Engineer and also very less Weightage is given to Factors like Education Qualification and Work Experience (If they chose to go with the Orginal Data). It is like a goto Tool specially meant for Recruiters and Talent Acquitions so that they can handle Large number of Applicants in an organised way.

Machine Learning Approach

How much Coding do you require in order to run PyRecruiter?

This Package specially focuses on Recruiter's ease that is why it has very few lines of code in order to run our default model (Actually around 3-4 lines to be honest). If you are familiar with Python and it's Libraries, you can definitely use it to make more accurate or add your own optimised models but the Default model has been well optimized and is more than enough for people with no background knowledge in Machine learning or Data science.

How it Works?

It is a Two Step Process where the Recruiter can take Candidate Information through Microsoft Forms Template for which the link would be given in below in the Readme File. Candidates would fill the form and the response would be directed to the Recruiter's response excel sheet.

Let's Understand how Recruiters can Play around with the Microsoft Template and what are the Guidelines?

Link to the Shareable Template : Microsoft Forms Template

Recruiters are requested to Understand what how would this Form do and upto what limit they can change the Form according to the Job Description.Note: Recruiters are requested to Duplicate the Same Template and Recruiters can change the Question content based on the JD but the whole Meaning and the Options should not Change

Let's Take an Example and compare a Random Job Description and see how a Recruiter can use the Information from JD to the Form

This Package also gives Recruiters an additional option to know if the Candidate is looking for a Domain change towards ML/AI.

Even though the Experience part here is not given in this JD but PyRecruiter gives Recruiters to get Experience info from Candidate through 3 Types which can be Low , Medium and High. Recruiters can Change the Experience Distribution in the Question but the important thing is to keep the importance from Low to High form.

This Package gives Recruiters 3 Options for adding Programming Languages. For this example it can be Python, R and Java. This information can be changed but the Options should not be changed

It also have options for adding 4 Required Skills/Libraries/Frameworks from the JD. In this example it can be Numpy , Tensorflow , Ggplot etc. as Python and R is required as a Language.

If there are Required Skills then Candidates should have Applied Projects with regards to the same skills. This Package accepts 4 types which should be the Number of Projects the Candidates has with respect to the Number of Skills.

This Package also gives an important additional skill which is the Number of Story-Telling Projects a Candidate has, which are 3 at Max (Currenty).

This Package also gives another important additional skill which is the Number of Domain Centric Projects a Candidate has, which are 3 at Max (Currenty).

This Package also gives asks for the Candidate's Education Qualification which are Bachelors , Masters and PHD. NOTE: This info can be changed but the Order of Importance and the Options should be Same like 3 would be the Highest one and followed by 2 and 1.

This Package gives slight Weightage to Good Certifications like the Test Based Certificates and Less Importance to Course Based and No Certificates.

How can a Recruiter use PyRecruiter?

Codes

Recruiters can take that Response excel sheet and go to their Command line and do following Codes:-

pip install pyrecruiter as pr

Here the Recruiter can Install the PyRecruiter package from PyPi

model = pr.Model()

This command would initiate the Model class that is in the .py file.

model.train_model()

Here the user gets an option by including dataset parameter which gives Recruiters an option to include either the dataset = 'orig_data (Default) or dataset = experience Data which is biased towards high Experience.

model.load_test_data(data_path='New Microsoft Excel Worksheet.xlsx')

Here Recruiters can provide the path to the Response excel sheet which they get through Microsoft forms

model.evaluate()

Now they can see the Results in the command line as well as they get an Excel Sheet containing the Names and Email IDs of all Candidates with their Probability of Getting shortlisted as well as the Predicted Result.

How can a Job Seeker or a Contributer who is familiar with Python can use PyRecruiter?

pip install pyrecruiter as pr

Here the User can Install the PyRecruiter package from PyPi

Model_data = pr.Model()

This command would initiate the Model class that is in the .py file.

Model_data = pr.Model(dataset='experience_data')

Here the Users gets an option by including dataset parameter which gives Recruiters an option to include either the dataset = 'orig_data (Default) or dataset = experience Data which is biased towards high Experience.

Now the Contributers can create their own model with the data they want to train on and can imporove the Accuracy or include new features through Feature Engineering.

model.load_test_data(data_path='New Microsoft Excel Worksheet.xlsx')

Here Users can provide the path to the Response excel sheet which they get through Microsoft forms

model.evaluate()

Now they can see the Results in the command line as well as they get an Excel Sheet containing the Names and Email IDs of all Candidates with their Probability of Getting shortlisted as well as the Predicted Result.

Results

This Package focuses on Minimum code and the Right Parameters for Shortlisting Candidates apart from the Traditional Parameters like Education Qualification and Work Experience.

Change Log

0.0.1 (14/11/2020)

  • First Release

0.0.2 (15/11/2020)

  • Second Release

0.0.3 (15/11/2020)

  • Third Release

0.0.4 (15/11/2020)

  • Fourth Release

0.0.5 (16/11/2020)

  • Fifth Release

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pyrecruiter-0.0.5.tar.gz (7.2 kB view details)

Uploaded Source

File details

Details for the file pyrecruiter-0.0.5.tar.gz.

File metadata

  • Download URL: pyrecruiter-0.0.5.tar.gz
  • Upload date:
  • Size: 7.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.2.0 pkginfo/1.5.0.1 requests/2.24.0 setuptools/49.2.0.post20200714 requests-toolbelt/0.9.1 tqdm/4.47.0 CPython/3.7.7

File hashes

Hashes for pyrecruiter-0.0.5.tar.gz
Algorithm Hash digest
SHA256 f75636685e591b7caf26506170094a85870d25b58be7e074b50ceb0eba10a2ac
MD5 d8e282a585d840623c1176c881bb6b91
BLAKE2b-256 4dc6e4cbd29bd09da3d16c790f3976da7dd73682b826ae11490660b1ab2939eb

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page