SHL Assessment Recommendation Engine

Project Overview

This project is an end-to-end, content-based recommendation system designed to match a user's hiring requirement (e.g., job role, skill, or level) to the most relevant assessments within the SHL product catalog.

The solution is implemented as a simple web application using Streamlit for a live, interactive demonstration.

🛠️ Technology and Methodology

Component	Technology/Method	Purpose
Backend Logic	Python, scikit-learn	Core recommendation logic.
Recommendation Algorithm	TF-IDF Vectorization and Cosine Similarity	Converts assessment text and job requirements into numerical vectors and calculates similarity to provide the best match.
Data Source	shl_catalog_.csv (Internal SHL Product Data)	Contains the features (Test Name, Description, Job Levels) used for vectorization.
Frontend/Deployment	Streamlit	Provides a simple, interactive User Interface (UI) and allows for quick cloud deployment.

Recommendation Logic Explained

The engine operates on Content-Based Filtering.

1. Data Combination: The three key textual fields (Individual Test Solutions, Description, and Job Levels) are concatenated into a single feature space called combined_text. This ensures the model captures keywords from all relevant areas.
2. Vectorization (TF-IDF): The combined_text for all 377 assessments is transformed into numerical vectors using TF-IDF (Term Frequency-Inverse Document Frequency). TF-IDF gives higher weight to rare, meaningful keywords (e.g., 'Python', 'Director') over common ones (e.g., 'test', 'measures').
3. Similarity Calculation: When a user enters a query (e.g., "Python developer"), it is also converted into a TF-IDF vector. Cosine Similarity (a measure of the angle between two vectors) is used to find the assessments whose vectors are closest to the query vector.
4. Output: The top 5 closest matches are returned to the user.

How to Run Locally

1. Clone the repository:

   git clone [YOUR_REPO_URL]
   cd shl-assessment-recommender

2. Install dependencies:

   pip install -r requirements.txt

3. Run the Streamlit app:

   streamlit run app.py

🔗 Live Demo URL

https://shlassessment-recommendation-engine-xyznl68gzkstuvyeg45g67.streamlit.app/

---

Created by Sourav Pati for the SHL Research Intern Assessment.