Home
Manage Projects
Students
About us
Guide
Available Projects
Finished Projects
Info
Newspaper
Contact
sign in
sign up
Home
Manage Projects
Students
About Us
Available Projects
Finished Projects
Info
Newspaper
Contact
sign in
sign up
Anomaly detection in iOT - SmartCampus
Data Science
Project Guide :
Mayya Lihovidov
Development :
Start :
2024-03-31
Finish :
2024-09-24
Hebrew Year :
תשפד
Semesters :
2nd & 3rd
Description
Smart Campus Data Monitorings Project Title: Smart Campus Data Monitorings Project Brief description: Smart Campus AI project contains three main sub-projects: 1. Data extraction and procedure storage 2. Data Exploration and Analysis 3. Prediction and Anomaly Detection Data Extraction: Currently data is stored among multiple tables within the relational database, and can be extracted and manipulated by SQL. The best case scenario is to store the procedures in SQL so that any group can extract ready for the exploration dataset. Alternatively, it might be one time written query in order to extract CSV for the future exploration in Python. Data Exploration and Analysis: At this stage data is manipulated in Python in order to detect the correlations, dependencies, data inconsistencies, and more within the standard EDA process. Prediction and Anomaly Detection Data should be classified in order to automatically detect patterns within different timeframes for each and every parameter as well as the combination of the parameters. Prediction of the normal expected values will help to detect outliers, signalizing about the following possible problems: Applications of the Solution The solution is going to be used within HIT Smart Campus project for data analytics for Sensor based data to manage the following problems: Overconsumption of the electricity Overconsumption of the water Unusually big/small number of students per timeframe Overpopulated classes Security breaches Time frame 2 Semesters project 1. 4 weeks to create and validate the dataset 2. 6 weeks for the data exploration 3. 6 weeks to train and fine-tune the models Technologies used within the Project Python SQL Pandas Number of participants: 2-3 groups of 4-7 students each. A project designed for students of computer science / applied mathematics. For any questions about the project, I will be available by phone: 050-5714100 Regards, Mark Israel markisr@walla.co.il, marki@hit.ac.il
Emphasis in project execution
The project is has cooperation with the industry and combines meeting deadlines while being creative and focused on the task
Status:
Shown in Available Projects
THE PROJECT IS AT FULL CAPACITY
Create New Student Profile + Register to this Project
I have a question