School of Business

Permanent URI for this communityhttps://rda.sliit.lk/handle/123456789/4207

Browse

Filters

2016 - 201922020 - 20213

Advanced Search

Filter by

Settings

Subject: search.filters.subject.Natural Language Processing

Search Results

Now showing 1 - 5 of 5
  • Thumbnail Image
    PublicationOpen Access
    Simplifying Law Statements Using Natural Language Processing
    (SLIIT, 2016-11-16) Dharmasiri, N; Gunathilake, B; Pathirana, u; Senevirathne, S; Nugaliyadde, A; Thelijjagoda, S
    Understanding the law statements for general public is evidently complex. The research derives a computational solution on reducing the complexity of the law statements. Given a law statement, the research will use both wordnet and “LawNet” to create a simpler meaning. The research will focus on information extraction, information retrieval, question analysis and answer generation techniques to derive better meaning of law statements. The law statement will be treated as a question and the “LawNet” and wordnet will be used in as information extraction points. The law statement will be analyzed as a question; more information will be retrieved through the wordnet and “LawNet”. This process mostly acts similar to a search engine’s process. The results provide on average 80% accuracy for a 1500 dataset.
  • Thumbnail Image
    PublicationEmbargo
    Automate Traditional Interviewing Process Using Natural Language Processing and Machine Learning
    (IEEE, 2021-04-02) Senarathne, P; Silva, M; Methmini, A; Kavinda, D; Thelijjagoda, S
    Nowadays, almost everything is equipped with technology. People can save time by using modern day technological applications in the most convenient way. Smart Interviewing System is one such software/tool which automates the traditional interviewing process using modern Natural Language Processing techniques and deep learning applications. The system will be mainly beneficial for interviewers and HR management employees working for different organizations who conduct technology related interviews. The system works with human voice and writing patterns. The system converts human language into system understandable text-based inputs, and these are used as inputs in the automated interviewing process. The system then checks the accuracy of the answers which candidates provided on the both oral interviews/ technical interviews and written tests. Later, the system automatically predicts scores for each answer using concepts of the deep learning. Interviewers can reduce the effort that they have to put in for selecting the most suitable candidates who are qualified enough to work with their organization. SIS is developed based on modern DL and NLP concepts using Python programming language alongside with ReactJS Framework. This system checking and evaluating candidate more accurately in every stage of the interview using advance evaluation parameters than human oriented evaluations. Above process lead system to find more human errors which critically can be affected to future of the organizations. Because of that, it can be led organizations to find best human resources comparing to the traditional interviewing process by sacrificing less time and effort.
  • Thumbnail Image
    PublicationEmbargo
    Plagiarism Detection Tool for Sinhala Language with Internet Resources Using Natural Language Processing
    (IEEE, 2021-08-11) Rajamanthri, L; Thelijjagoda, S
    With the digitalization of text through the World Wide Web, plagiarism turned into a crucial problem and a way to detect plagiarism became an essential component. Even though there are many plagiarism detection systems, applying in the world by considering other languages, for Sinhala it’s only a few, and it senses a vacuum in the domain. With the significant improvement of availability in Sinhala text on WWW, still, there is no system to detect plagiarism for these Sinhala documents by comparing resources on the internet. The purpose of this research is to address and overcome the above-mentioned gap while introducing a plagiarism detection system for the Sinhala language, by using internet resources. In the process to obtain the outcome, text pre-processing, Google searching and similarity comparison using Jaccard coefficient were the steps followed. As the final product, a Sinhala plagiarism detection tool was developed with 88% accuracy. The outcome of the work will be a support for the lectures, teachers, authors, students who are using Sinhala as their literal language.
  • Thumbnail Image
    PublicationEmbargo
    Sinhala language decoder
    (IEEE, 2018-10-02) Vidanaralage, A. J; Illangakoon, A. U; Sumanaweera, S. Y; Pavithra, C; Thelijjagoda, S
    The purpose of this paper is to present a Sinhala and English bilingual translator to overcome language barrier which prevails in Sri Lankan organizations when collaborating business across international boundaries. First phase of the research involves statistically analyzing relationships and validating the semantic meaning between Sinhala and English sentences. Next phase is to develop Singlish transliteration approach to render Sinhala words typed using English. Final phase of the study involves recognizing handwritten Sinhala characters used in modern literature within an acceptable accuracy. The anticipated outcome is to improve communication effectiveness when doing business and to minimize time, cost, and human errors caused by a human translator.
  • Thumbnail Image
    PublicationEmbargo
    Arunalu: Learning Ecosystem to Overcome Sinhala Reading Weakness due to Dyslexia
    (2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Sandathara, L.; Tissera, S.; Sathsarani, R.; Hapuarachchi, H.; Thelijjagoda, S.
    Dyslexia is an impairment in ability in reading. People having Dyslexia has difficulties in identifying specific letters and words and identifying speech sounds and decoding the letters which leads to difficulties in comprehension, spelling and writing. Dyslexia may severely affects language development and impacts reading and other language based improvement and functioning. “ARUNALU: Learning ecosystem to overcome reading disabilities in Sinhala language due to Dyslexia” has been proposed as a multi-sensory mobile solution, in native language of Sri Lanka (Sinhala) and with effective screening and intervention methodologies recommended by health professionals. Objective is to deliver, a phonological support to enhance reading skills of dyslexic children by providing a machine learning based automated screening and intervention mobile solution. Through these reading environments, there's a reward system in intervention process to encourage the user, and also users and respective parties can analyze user's progress. The proposed system is mainly based on Voice recognition, Natural Language Processing, Machine Learning and Deep Learning concepts collaborating with reading and gaming environments. Core Objective of the proposed system is to come up with a better and effective screening and intervention methodologies for early identification of Dyslexia and provide phonological training to overcome Sinhala reading difficulties due to Dyslexia in a user friendly manner.