Wahidul Alam Riyad's Projects
In implementing the ant colony optimization algorithm, the developer has used 500 ants and 8 US-based cities to calculate the shortest route for the TSP. The developer found out that the number of ants does not affect finding the shortest route. They also observed that the number of cities does not increase the time to calculate the shortest route. In ACO, the shortest route's distance is 6047 miles, but every time the program has generated the ant, which finds the best path is always different. The course is still different, and even the distance remains the same.
Conducted research and developed a system under Dr Booma Poolan Marikannan on provisional analysis for obesity issues using numerous data mining techniques by using a past medical dataset from the Kaggle. Executed the project using tools such as PyCaret, Pandas, NumPy, Matplotlib, Seaborn, Scikit-Learn, and Pickle, and evaluated the classification models to classify obesity based on the value of BMI by using Classification Report and Confusion Matrix. Achievement - Implemented supervised machine learning techniques, such as Quadratic Discriminant Analysis, K-Nearest Neighbour, and Random Forest, with an accuracy of 91% to forecast customers' profitability based on consumer products.
In the brute force algorithm, the program will go through all the permutations of starting a trip from one of the cities, visiting each city once, and returning to the first cities. It shows all the possible shortest routes with the same distance out of all the permutations. The downside of this algorithm is that increasing the number of cities increases the overall time to find the best shortest route, inefficient for the TSP. The shortest distance that the authors found from this algorithm is 6047 miles, which is the same as the ACO algorithm.
Conducted research and developed a system under Dr Chris Walshaw by implementing numerous data visualisation techniques to showcase the customers’ visits to the store using the ChrisCo Company dataset. Executed the project using tools such as HoloViews, Pandas, NumPy, Matplotlib, and Seaborn, and applied Exploratory Data Analysis to find crucial patterns of customers’ visits to the ChrisCo company by categorising them into high, medium and low. Achievement - Implemented Line Plot, Bar Chart, Scatter Plot, Heatmap Plot, Pearson Coefficient Scatter Plot, Seasonality Graph, Line Plot Interactive and Heatmap Interactive Plot to visualise all the insightful data of the customers’ visits.
Conducted research and developed a system under Dr Jixin Ma on the comparison of numerous classification models to predict coronary heart disease using past medical data from the UCI Machine Learning Repository. Executed the project using tools such as Pandas, NumPy, Matplotlib, Seaborn, and Scikit-Learn, and evaluated the classification models...
DressMe Clothing Marketplace (DCM) System has been created for DressMe Clothing Sdn. Bhd. Its main purpose is to place an order, view order, add a new cloth, search for a type of clothing, move back and forth between the clothes that are for sale. Select clothes to order, checkout order, move back and forth between orders (prioritize orders that require delivery), modify an order, delete an order. Doubly linked lists have been implemented using a C++ programming language to build this system.
This project is ‘Dyscalculia Games,’ an online math learning platform that provides various math games for users suffering from dyscalculia to aid their treatment. The website is developed using the programming language PHP in both front end and back end aspect. The users intend to target is children aged 5-6 who are suffering from dyscalculia. The website contains various math games covering basic operations (addition, subtraction, etc.). It will store the user’s score as well for score tracking purposes and a high score system.
Developed the Food Vision Big model using TensorFlow, surpassing the performance of the 2016 DeepFood CNN model with an accuracy of 80.2% on the Food101 dataset comprising 101,000 images. Implemented advanced training techniques including prefetching and mixed precision training, reducing model training time to approximately 20 minutes compared...
In implementing the hill-climbing algorithm, the program first generated a distance based on the result, and it will loop back to previous routes. Then as it connects back, the next course swaps randomly two cities, and it loops again for the shortest path. It will stay on the same route for more iteration until it reaches the shortest path. The entire program will end when it reaches 100 iterations as the developer has set the potential maximum. The shortest distance for this algorithm is 6977 miles.
This capstone project course will showcase to find an apartment in Manhattan with the following conditions. Apartment with min 2 bedrooms. Monthly rent not to exceed US$7000 / month. Located within walking distance (<=1.0 mile, 1.6 km) from a subway metro station in Manhattan. Venues and amenities that are available within my current residence.
Pokédex
Library Management System (LMS) is a software created by Java to manage the catalog of a library. The LMS helps to keep the records of whole transactions of the books available in the library. The LMS can log in, Register Users, Add Books, Register New Students, Issue Book, Return Book, shows Statistics of Issued Book and Returned Book, shows a list of Registered Students and All Books. It has great UI, such as a navigation bar, application theme, and many more.
The objective is to design an efficient automatic authorized vehicle identification system by using the vehicle number plate. The system is implemented on the entrance for security control of a highly restricted area like military zones or area around top government offices e.g. Parliament, Supreme Court etc. The developed system first detects the vehicle and then captures the vehicle image. Vehicle number plate region is extracted using the image segmentation in an image. Optical character recognition technology is used for character recognition. The resulting data is then used to compare with the records on a database so as to come up with the specific information like the owner of the vehicle, place of registration, address, etc.
The project was to produce a movie ticketing system using Visual Basic .Net, enabling the user to book a ticket for available movies. Users can check availability for a movie and make a reservation for a specific date and seat. There are three types of seats available: Standard, Gold, and Premium. Customers can select seats depending on personal preference. Each type has its price. Standard is for RM 15, and Gold is for RM 20, and Premium is for RM 45. Regular staff can make new reservations, edit reservations, and enter payments. Only managers can access the reports while the regular staff cannot access the reports.
Conducted research and developed a system under Dr Stef Garasto on the comparison of numerous machine learning models to solve a multi-task classification problem using an existing dataset for the Amazon Product Reviews. Executed the project using modules such as NLTK, Tensorflow, Pandas, NumPy, Matplotlib, Seaborn, and Scikit-Learn, and evaluated the classification models to predict the ratings of the Amazon Reviews and Product Categories. Achievement - Implemented supervised machine learning techniques such as Logistic Regression, Support Vector Machine, Naive Bayes and Random Forest to conduct this research and achieved an accuracy of 63% using TFIDF Vectorisation.
The nearest neighbor algorithm gets the shortest route from the initial course whenever generating the result. It will always check for the journey to reduce one city from the way, which is the shortest route, and then it will run again for the remaining cities until it gets the best result. It follows the pattern until there are no remaining cities. The distance of the shortest route for this algorithm is 7184 miles.
ML x Health consists of Bayesian ML, Representation Learning, Graph Neural Networks, Computer Vision, Knowledge Graphs, Knowledge-Aware ML, Symbolic Reasoning, Neuro-Symbolic AI, EHR, Imaging, Genomics, and Multi-Omics. ML x Finance consists of Gaussian Processes, Time Series, Multi-Lingual NLP, Reinforcement Learning, Building Market Simulators, Trading, Asset Management, Financial Inclusion, ESG, Emerging Risks, and Economic Prosperity. Achievement - Completing all tasks before the deadline with 100% output and achieving the Certificate of Completion. Health and Finance Tracks were crucial in completing the MSc Data Science Final Year Dissertation.
The Pharmacy Management System (PMS) helps to perform the administrative activities for the Pharmacy shop. With this new system, the pharmacy staff will find it more natural to complete all the necessary operations for their daily business transactions. Admin users will be responsible for adding, delete, searching, and modifying records related to medicine. Customers can search for a particular medicine availability and the respective details regarding that medicine. The developer created the application using C++ and Object-Oriented Concepts to achieve all the required functionalities.
Developed an advanced question-answering system utilising the OpenAI, Pinecone, and LangChain (OPL) stack, enabling dynamic information retrieval from nonpublic or recent documents not covered in the model's training data. Implemented a retrieval-augmented generation approach using embeddings to efficiently process and query large text corpora...
In this project, students must design a Retail Order Management System using Java for small retailers. And there are two different stockholders, admin, and customer. Admin can manage customers, manage products, and manage orders. Based on the system requirements, the customer can only manage orders. The developer learns the fundamentals of object-oriented programmings, such as inheritance, polymorphism, encapsulation, and abstraction. The developer got the chance to differentiate between a variety of data types and its unique features, for example, Vectors and ArrayList, then implemented it to their project as well as developer learn how to allocate resources effectively (memory), or how to design an optimal algorithm (timely efficient).
sending-emails-with-python
The simulated annealing algorithm started with a random route and got a random number for the route. When the number is smaller and equal to the probability function, it will proceed to the adjacent route. If the adjoining route is larger than the current route, it will still proceed to the next route until it gets the shortest adjacent route. If the random number is larger than the probability of function, it will stay in the same route until it reaches the shortest route to follow by the iteration. The program will end when it is no more probability of function, and a result shows the shortest route. The shortest distance that the authors found from this algorithm is 6994 miles.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent