Research Papers - Dept of Information Technology

Permanent URI for this collectionhttps://rda.sliit.lk/handle/123456789/593

Browse

Filters

2018 - 201922020 - 20235

Advanced Search

Filter by

Settings

Has files: YesSubject: search.filters.subject.machine learning

Search Results

Now showing 1 - 7 of 7
  • Thumbnail Image
    PublicationEmbargo
    Cloud-based Salesman-Bot for Ontology-based Negotiation
    (IEEE, 2023-04-06) Fernando, A; Rahubedda, T; Jayasinghe, B; Mallikahewa, S; Hettiarachchi, O; Rajapaksha, S
    We have proposed a cloud-based ChatBot (Salesman-Bot) approach to handling multiple negotiation scenarios in a supermarket environment. The web application is a simple interface that can be implemented on a single standalone device or interacted with through a mobile phone. The Salesman-Bot responds both via text and speech. By introducing a Salesman-Bot, efficient negotiation, with quick preferences and suggestions can be provided. A new architecture proposed to operate the Salesman-Bot together with Google APIs and libraries such as Natural Language AI, Vision AI, Speech to Text API, Text to Speech API and Machine Learning using TensorFlow. The application also uses the Google Cloud Platform with related services such as Google App Engine. The goal is to make ChatBots more efficient in negotiating in different business scenarios. This paper presents the work carried out with ontology and machine learning in a cloud-based environment to handle multiple negotiation scenarios based on a negotiation hierarchy. It also proposes the opportunities and drawbacks of such a system.
  • Thumbnail Image
    PublicationOpen Access
    Hashtag Generator and Content Authenticator
    (researchgate.net, 2018-01) Yapa Abeywardena, K; Ginige, A. R; Herath, N; Somarathne, H; Thennakoon, T. M. N. S
    In the recent past, Online Marketing applications have been a focus of research. But still there are enormous challenges on the accuracy and authenticity of the content posted through social media. And if the social media business platforms are considered, majority of the users who try to add a market value to their own product face the problem of not getting enough attention from their target audience. The purpose of this research is to develop a safe and efficient trending hashtag generating application solution for social media business users which generates trending and relevant hashtags for user content in order to get a broad reach of target audience, automatically generates a meaningful caption to their relevant posts and guarantees the authenticity of the product at the same time. The user content is analyzed and filters the important keywords, generates a meaningful caption, suggest related trending keywords and generates trending hashtags to get the required reach for online marketers. Additionally, the marketing products’ content authentication is ensured. The application uses Natural Language Processing, Machine Learning, API technologies, Java and Python technologies. A unique database is assigned to users which contains rankings for each user. The target audience who engages in buying products get to know about the status of the sellers with respect to authenticity of the content. It is believed that the application provides a promising solution to existing audience reach problems of online marketers and buyers. The significance of this system is to help marketers and buyers to engage in online buying and selling with much effective, reliable and safer ways. This mitigate the vulnerability of bad social media marketing influences and helps to establish a safe and reliable online marketing practice to make both sellers and buyers happy. This paper provides a brief description on how to perform an organized online marketing discipline via the Trending Hashtag Generator & Image Authenticator application.
  • Thumbnail Image
    PublicationEmbargo
    Automatic Sinhala News Classification Approach for News Platforms
    (Institute of Electrical and Electronics Engineers Inc., 2020-12-18) Kirindage, G; Godewithana, N
    Because of generating various news articles in large scale, online sources moved into an automatic categorization mechanism. This research has been conducted using LDA topic modeling approach and using other classification algorithms to establish a news categorization solution. Sinhala news websites have only few news categories and do not have any relationships or hierarchies between the categories. Therefore, some users require to search manually and find the necessary articles which are in those categories. Purpose of this study is to build a news categorization model with categorization hierarchies for Sinhala news articles. The goals of the models are to identify the most suitable news category for a related news article and develop hierarchies using generated news categories and assign the news articles according to the hierarchical structure. The final experiments and evaluations show that the solution performs well to solve the automatic categorization problem in Sinhala news platforms.
  • Thumbnail Image
    PublicationOpen Access
    A predictive model for paediatric autism screening
    (SAGE Publications, 2020-12) Wingfield, B; Miller, S; Yogarajah, P; Kerr, D; Gardiner, B; Seneviratne, S; Samarasinghe, P; Coleman, S
    Autism spectrum disorder is an umbrella term for a group of neurodevelopmental disorders that is associated with impairments to social interaction, communication, and behaviour. Typically, autism spectrum disorder is first detected with a screening tool (e.g. modified checklist for autism in toddlers). However, the interpretation of autism spectrum disorder behavioural symptoms varies across cultures: the sensitivity of modified checklist for autism in toddlers is as low as 25 per cent in Sri Lanka. A culturally sensitive screening tool called pictorial autism assessment schedule has overcome this problem. Low- and middle-income countries have a shortage of mental health specialists, which is a key barrier for obtaining an early autism spectrum disorder diagnosis. Early identification of autism spectrum disorder enables intervention before atypical patterns of behaviour and brain function become established. This article proposes a culturally sensitive autism spectrum disorder screening mobile application. The proposed application embeds an intelligent machine learning model and uses a clinically validated symptom checklist to monitor and detect autism spectrum disorder in low- and middle-income countries for the first time. Machine learning models were trained on clinical pictorial autism assessment schedule data and their predictive performance was evaluated, which demonstrated that the random forest was the optimal classifier (area under the receiver operating characteristic (0.98)) for embedding into the mobile screening tool. In addition, feature selection demonstrated that many pictorial autism assessment schedule questions are redundant and can be removed to optimise the screening process.
  • Thumbnail Image
    PublicationEmbargo
    Analyzing Payment Behaviors And Introducing An Optimal Credit Limit
    (2019 1st International Conference on Advancements in Computing (ICAC), SLIIT, 2019-12-05) Bandara, H.M.M.T.; Samarasinghe, D.P.; Manchanayake, S.M.A.M.
    Identifying an optimal credit limit plays a vital role in telecommunication industry as the credit limit given to customers is influence on the market, revenue stabilization and customer retention. Most of the time service providers offer a fixed credit limit for customers which may cause customer dissatisfaction and loss of potential revenue. Therefore, it is essential to determine an optimal credit limit that maintains customer satisfaction while stabilizing the company revenue. Clustering algorithms were used to group customers with similar payment and usage behaviors. Then the optimal credit limit derived for each cluster is applicable to all the customers within the cluster. In order to identify the most suitable clustering algorithm, cluster validation statistics namely, Silhouette and Dunn indexes were used in this research. Based on the scores generated from these statistics KMeans algorithm was chosen. Furthermore, the quality of the KMeans clustering was evaluated using Silhouette score and the Elbow method. The optimal number of clusters are identified by those validation statistics. The significance of this approach is that the optimal credit limits generated by these clustering models suit dynamic behaviors of the customer which in turn increases customer satisfaction while contributing to reducing customer churn and potential loss of revenue.
  • Thumbnail Image
    PublicationEmbargo
    Facial Emotion Prediction through Action Units and Deep Learning
    (2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Nadeeshani, M.; Jayaweera, A.; Samarasinghe, P.
    With the recent advancements in deep learning techniques, attention has been given to training and testing facial emotions through highly complex deep learning systems. In this paper we apply machine learning techniques which require less resources to produce comparable results for emotion prediction. As the underlying technique for the emotion prediction in this research is based on clinically recognized Facial Action Coding System (FACS), a further analysis is given on the contribution of each of the Action Units (AUs) for the predicted emotion. This analysis would complement, strengthen and be a main resource for addressing many different health issues related to facial muscle movements.
  • Thumbnail Image
    PublicationEmbargo
    AI Base E-Learning Solution to Motivate and Assist Primary School Students
    (2021 3rd International Conference on Advancements in Computing (ICAC), SLIIT, 2021-12-09) Silva, P.H.D.D.; Sudasinghe, S.A.V.D.; Hansika, P.D.U.; Gamage, M.P.; Gamage, M.P.A.W.
    E-learning is a form of providing education by using electronic devices. Lack of proper mechanisms for encouraging and assisting students are key issues faced by many students in an e-learning environment. The ‘Vidu Mithuru’ is a question-based e-learning application which has been developed as a solution to overcome these problems. This mobile application will auto generate and categorize the questions, evaluate the answers and track the performance while providing motivational quotes by detecting the emotions of the student. This mobile application is based on Neural Networks, Natural Language Processing and Machine Learning concepts. In order to developing this application, the information provided by the primary education professionals was used to comply with the standards. The core objective of the proposed solution is to track the performance level and assist the students to improve in their studies while keeping them motivated. The trained Machine Learning models have achieved the accuracy of 75%, 78%, 99% and 86% for question categorization model, speech emotion detection model, facial emotion detection model and model to evaluate answers as respectively. We have received favorable responses as the results after testing the developed ‘Vidu Mithuru’ mobile application