International Journal of Computer (IJC)

Fuzzy Logic Based Dam Water Shutter Control System by Using Water Level and Rainfall Condition in Raining Season

Nay Thazin Htun, Moh Moh Nyunt Aung — Sun, 18 Jul 2021 15:14:24 +0000

A robust water shutter management system ensures that the water does not overflow and destroy or damage the dam. During the rainy season, care must be taken when dams conserve water, if the reservoir volume is too high, the risk of dam failure may be increased. So water level control is a special matter in the rainy season. Fuzzy logic sets provide better control than binary logic-based methods because they are used to determine the meaning of qualitative values for controller inputs and outputs, such as small, medium, and large control actions. This system used the fuzzy logic control theory in the water shutter management system to get smoothness motor control values of small, very small, medium, large, and very large. The system uses ultrasonic sensors to detect water levels, rain sensors to detect rain, and fuzzy logic controls to control the PWM duty cycle to the shutter gate motor driver circuit based on the detection of these two sensors. This control strategy is implemented with Arduino Uno.

Recognition of West African Indigenous Fruits using a Convolutional Neural Network Model

Wed, 04 Aug 2021 07:44:53 +0000

The. Fruit recognition involves the extraction and processing of relevant features from fruit images in order to deduce the categories of that fruit. Due to its importance to human health and sustainability, various systems exist for recognition of fruits, although none exist for recognition of west Africa's indigenous fruits. This research developed a fruit recognition system using a convolutional neural network (CNN) based model. Five west Africa indigenous fruits were selected, while “images were directly used as input to CNN based model of (3 convolutional layers, 3 max pooling layers and 1 fully connected layer) for training and recognition without features extraction process. The study further presents a transfer learning on visual geometry group 16 and ResNet models for result comparison. Using the optimal training set, the proposed CNN based model produced a recognition rate of 96%.

Online Laundry Management System

Thu, 04 Nov 2021 20:03:05 +0000

This study presents the automation of an online laundry management system (OLMS) for laundry organizations. Laundry firms usually have the challenges of keeping detailed records of customers clothing leading to disappointments on the side of customers. Issues arising include customer clothes mix-ups and untimely retrieval of clothes collected in relation to their owners. This system helps the users track progress on their clothing items, fixes date for collection or arranges drop-offs and communicates directly with business operators. Also, customers’ information remain available at all times as it is retained within the system. Each customer is assigned a unique ID on registration to avoid contrasting information. The implementation tools include PHP, JavaScript, HTML, MySQL, visual studio, WAMP server and a web browser. This solution brings ease to operating the business and controlling work flow; from managing customer information to managing service requests/orders as well as managing service rendition. The design also has a unique and user friendly interface. This affords the users and providers of the service an opportunity to enjoy seamless operations.

Performance Analysis of Machine Learning Models for Sales Forecast

Omogbhemhe Izah Mike, Odegua Rising — Sat, 27 Nov 2021 09:15:19 +0000

Many supermarkets today do not have a strong forecast of their yearly sales. This is mostly due to the lack of the skills, resources and knowledge to make sales estimation. At best, most supermarket and chain store use adhoc tools and processes to analyze and predict sales for the coming year. The use of traditional statistical method to forecast supermarket sales has met a lot of challenges unaddressed and mostly results in the creation of predictive models that perform poorly. The era of big data coupled with access to massive compute power has made machine learning model the best for sales forecast. In this paper, we investigated the forecasting of sales with three machine learning algorithms and compare their predictive ability. Three different methods used are K-Nearest Neighbor, Gradient Boosting and Random forest. The data used to train the machine learning models are data provided by Data Science Nigeria on the Zindi platform, the data were collected from a supermarket chain called “Chukwudi Supermarkets”. The results show that the Random Forest algorithm performs slightly better than the other two models, we saw that Gradient Boosting models were prone to over-fitting easily and that K-Nearest Neighbor even though fast, performs poorest among the three.