http://dlib.iit.ac.lk/xmlui/handle/123456789/1617 2026-04-08T18:30:57Z 2026-04-08T18:30:57Z Yasiru, Dushan http://dlib.iit.ac.lk/xmlui/handle/123456789/2191 2024-06-05T04:15:01Z 2023-01-01T00:00:00Z

Tomato Leaf Disease Detection Using Machine Learning Yasiru, Dushan "Tomato cultivation in Sri Lanka faces challenges from unreliable disease identification systems, impacting crop yields. To address this, an innovative and user-friendly plant disease identification system is needed, utilizing advanced machine learning techniques to accurately detect various tomato leaf diseases, focusing on the Plateena variant. Overcoming funding and skill limitations is crucial for successful implementation."

2023-01-01T00:00:00Z Wattegama, Deshan http://dlib.iit.ac.lk/xmlui/handle/123456789/2190 2024-06-05T04:12:53Z 2023-01-01T00:00:00Z

“iCrops” – Intelligent Crop tracker and Recommender Wattegama, Deshan "Chemical fertilizer usage is big problem in Sri Lanka. Researchers say that the chronic kidney disease is also an effect of this chemical fertilizer usage. As traditional farming country Sri Lankan farmers were using organic fertilizer in the ancient history. But because of the increased requirements and due to the need of maximum profits farmers switched to the chemical fertilizer. To solve this problem, promotion of organic produce is a necessary thing. Organic products are not very much famous in Sri Lanka because of its price, trust issues and accessibility issues. “iCrops” application is proposed to solve this problem. This application creates a model between farmer, supermarket, and customer to promote organic product. Fresh product home delivery in Sri Lanka is still in its early stages. And the transportation cost is also quite high due to the economic situation in Sri Lanka. So this application can be used by the supermarkets to manage customer organic product orders. Thus, reducing cost and solving accessibility issues. To make this applications recommendations more trustworthy, explainable recommendations are generated. Most of the recommendation systems does not explain the generated recommendations. Explainable recommendations are a rising area of recommendation systems. Combining it with popular transformer models this application generates an explanation to win the trust of the user."

2023-01-01T00:00:00Z Perera, Dimalka http://dlib.iit.ac.lk/xmlui/handle/123456789/1701 2024-02-15T08:14:11Z 2023-01-01T00:00:00Z

Offensive Word Detection on the Images Intended to Mislead Bots Perera, Dimalka The research aims to detect offensive words on images written in the Sinhala language, particularly those covered with jargon characters. The author explains the problem domain, highlighting the prevalence of abusive language on social media and its manifestation in memes. The research defines the problem statement, emphasizing the need to address the loopholes utilized by "memers" to hide offensive content. The research aims to contribute to the body of knowledge by developing a robust system for detecting and flagging offensive words in Sinhala images. The research question and the novelty of the research in the domain are presented, along with the identified research gap and potential contributions. A system to detect crossed letters on images is proposed and designed and the system will also be able to extract the clean text from the image. Then an abusive language detection model will recognize the text as offensive or not offensive. A sequential CNN was implemented to detect the crossed letters and K-Nearest Neighbor, Random Forest and Support Vector Machine algorithms were used to detect offensive language in text format. Sequential CNN model trained on a limited dataset achieved an accuracy of 96% and KNearest Neighbor, Random Forest and Support Vector Machine algorithms have achieved an accuracy of 88% , 86% and 83% respectively.

2023-01-01T00:00:00Z Gunasekara, Sandani http://dlib.iit.ac.lk/xmlui/handle/123456789/1700 2024-02-15T08:09:37Z 2023-01-01T00:00:00Z

Detection of Mental Stress Level via code mixed Social media posts using Sentiment Analysis Gunasekara, Sandani "Detecting mental stress level using code-mixed social media posts lies in the unique linguistic and contextual challenges posed by the combination of multiple languages within a single post. Code-mixing, the phenomenon of blending languages, is prevalent in online communication, especially on social media platforms. This mixing of languages introduces complexities in natural language processing tasks, making it difficult to accurately detect and classify mental stress levels. The inclusion of multiple languages, variations in grammar and syntax, slang, and cultural references further complicate the analysis process. As a result, existing stress detection methods may not adequately handle code-mixed data, leading to reduced accuracy and effectiveness in identifying and understanding mental stress levels within this specific linguistic context."

2023-01-01T00:00:00Z