Smart recommender architecture for Internet of household items (IoHT) by using Edge and Fog Computing
With the assistance of IoT ,ML and hybrid computing architectures we will propose a smart reminder and recommender system that will help us in monitoring the kitchen grocery and refrigerating hot and cold items. Our proposed recommender IoHT architecture will able to recommen
2025-06-28 16:29:23 - Adil Khan
Smart recommender architecture for Internet of household items (IoHT) by using Edge and Fog Computing
Project Area of Specialization Computer ScienceProject SummaryWith the assistance of IoT ,ML and hybrid computing architectures we will propose a smart reminder and recommender system that will help us in monitoring the kitchen grocery and refrigerating hot and cold items. Our proposed recommender IoHT architecture will able to recommend and alert about available kitchen items status in real time and helps us in identifying items with low quantity .Moreover, the proposed IoHT model also consume minimum energy and have rapid response time which will increase the overall efficiency of a system . Finally, we will develop a smart android interface to display real time item values for visualization.
Project Objectives- To develop a layer based communication stack for Internet of Household Items (IoHT).
- To classify and design Hot and cold items based smart recommender policy.
- To gather and collect sensory dataset for IoHT.
- To design a hybrid computing model for IoHT.
- To deploy a smart recommender IoHT model in real-time.
%20by%20using%20Edge%20and%20Fog%20Computing'%20_1639954798.png)
In this project, our “smart recommender and reminder” system will help us to maintain the record of kitchen grocery items and Refrigerating items. We will attach different types of sensors on our kitchen items like sugar jar, oil bottle, rice container etc. Moreover, we will also use sensors for Refrigerating items like milk bottle, chicken spread, egg tray etc. These sensors will help us for reading the amount ofitems. After taking readings from the sensors, we will execute feature prioritization algorithm by using Raspberry pi (EDGE node) , Fog node and cloud services for analyzing useful and ordinary items. Then, we will make our IoHT dataset according to our daily routine and save it on the cloud server (AWS). We will use a wireless network for transferring data from IoT to the cloud. By taking advantage of Fog Computing, we will save data of useful items on it and the data of ordinary items on Cloud Computing. Then we will use different machine learning classifiers that help us in predicting and recommending things using our IoHT dataset. This prediction will recommend us to analyze about the importance of the items. Furthermore, a mobile application will be used to remind us the detail of each item, so that we can see what prediction is made by our smart system.
The fundamental benefit behind our idea is to prevent the wastage of food and its unnecessary stocking which ultimately dispose due to its validity. Moreover, it helps to recommend and identify particular kitchen item quantity. Furthermore, it can also monitor thekitchen stock and refill them timely without having to count them manually. Also, the fuel price is increasing day by day and the pollution is increasing at an alarming rate due to more fuel consumption .As a result, this smart recommender architecture will save our necessary time and energy and cost.
Technical Details of Final Deliverable%20by%20using%20Edge%20and%20Fog%20Computing'%20_1639954799.png)
Deliverable 1: For monitoring the record of our items, we will design sensory modules with kitchen and refrigerating items. These sensors will continuously read data from the items.
Deliverable 2: we will develop a sensory dataset of each item individually and save it on the cloud platform i.e. AWS.After saving data on cloud
Deliverable 3: we will prioritize and process items according to their features by using Hybrid computing such as Raspberry pi (Edge node) , Fog node and cloud platform.
Deliverable 4: Item’s data will be compute, process on the edge, fog and cloud computing mechanism for recommendation and prediction by using some machine learning techniques.
Deliverable 5: The recommended output along with real time kitchen and refrigerating items reminder notification will be shown on Android application.
Final Deliverable of the Project HW/SW integrated systemCore Industry ITOther Industries Others Core Technology Internet of Things (IoT)Other Technologies Artificial Intelligence(AI)Sustainable Development Goals Industry, Innovation and InfrastructureRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 71820 | |||
| Raspberry pie 4B (2GB) | Equipment | 2 | 15000 | 30000 |
| Sonar | Equipment | 10 | 350 | 3500 |
| Sonar (Water level) | Equipment | 5 | 650 | 3250 |
| Node MCU | Equipment | 10 | 800 | 8000 |
| Jumper wires | Miscellaneous | 10 | 150 | 1500 |
| Vero board | Equipment | 6 | 70 | 420 |
| IR sensor | Equipment | 2 | 200 | 400 |
| LCD | Equipment | 3 | 500 | 1500 |
| Arduino UNO | Equipment | 3 | 700 | 2100 |
| Arduino cable | Miscellaneous | 3 | 150 | 450 |
| Raspberry pi casing | Miscellaneous | 2 | 500 | 1000 |
| Raspberry pi power adopter | Equipment | 2 | 600 | 1200 |
| Sd card (32 GB) | Miscellaneous | 2 | 1500 | 3000 |
| PCB fabrication | Miscellaneous | 10 | 300 | 3000 |
| Buzzer / LED's/ resistors / buttons | Miscellaneous | 1 | 500 | 500 |
| Batteries (5 V, 3A) | Equipment | 12 | 1000 | 12000 |