Shareable Life | Enhancing Roommate Compatibility Detection through Machine Learning

Local Chapter Benha, Egypt

Coordinated byEgypt ,

Sandeep Sudesh Kumar

Status: Completed

Project Duration: 28 Apr 2023 - 01 Jun 2023

Open Source resources available from this project

Github Repository

Project background.

People who are studying away from their cities, especially students, often face difficulties in finding suitable roommates to share a flat. It’s not just about finding a good room or flat, but also about finding compatible roommates who share similar interests, lifestyles, and budgets. Machine learning can help solve this problem by asking clients a series of questions about their preferences and lifestyle and using a matching algorithm to identify the best potential roommates for them. This algorithm can be trained using a dataset of past matches to improve its accuracy and effectiveness.

The problem.

The complexity of finding suitable roommates can be daunting, and traditional methods like classified ads or social media may not be effective in finding the right match. Additionally, the process of manually assessing compatibility between potential roommates can be time-consuming and subjective, leading to suboptimal matches.

Project goals.

The primary goal of this project is to develop a machine-learning model that can accurately match potential roommates based on their preferences, lifestyle, and psychological profiles. The model should be efficient, streamlined, and user-friendly. The project aims to: - Collect data on potential roommates, including information on their lifestyle, food preferences, and other factors that may impact their compatibility with others. - Develop a machine learning model that uses the collected data to identify potential matches. - Train and evaluate the model using a dataset of past matches. - Develop a user-friendly app that incorporates the matching algorithm and enables users to input their preferences and receive recommendations for potential roommates.

Project plan.

  • Week 1

    Week 1: Data Collection
    During this week, the team will collect data on potential roommates, including information on their lifestyle, food preferences, and other factors that may impact their compatibility with others. The team will also consult with doctors to develop psychological questions that can be incorporated into the matching algorithm.

  • Week 2

    Week 2: Data Preprocessing and Feature Extraction
    During this week, the team will preprocess the collected data to ensure that it is suitable for use in the matching algorithm. They will also extract important features from the data to identify potential matches.

  • Week 3

    Week 3: Model Development In this week, the team will develop a machine learning model that uses the extracted features to identify potential matches. They will use a suitable algorithm, such as a decision tree or a neural network, to develop the model.

  • Week 4

    Week 4: Model Training and Evaluation During this week, the team will train the model using a dataset of past matches and evaluate its performance using a test dataset. They will fine-tune the model to improve its accuracy and effectiveness.

  • Week 5

    Week 5: App Development In the final week, the team will develop a user-friendly app that incorporates the matching algorithm and enables users to input their preferences and receive recommendations for potential roommates. The app will be designed to be easy to use and accessible to a wide range of users.

Learning outcomes.

The project will provide the following learning outcomes: Knowledge of machine learning algorithms and their applications in real-world problems. Experience in data collection, preprocessing, and feature extraction. Understanding of model development, training, and evaluation. Project management skills, including task planning, timeline management, and team coordination. Experience in developing a user-friendly app.

Share project on: