As Karunananda

A Parser for Sinhala Language - First Step Towards English to Sinhala Machine Translation

Sinhala language parser has been developed as the first step towards English to Sinhala natural language translation. The Sinhala parsing system comprises of a Sinhal a morphological analyzer that connects with three dictionaries, namely, base dictionary, rule dictionary and concept dictionary. The morphological analyzer of the parsing system implements inflection and derivation rules of the Sinhala grammar, and provides grammatical information of the words to be considered by the parser. The current version of the parser is capable of handling simple and complex sentences of Sinhala language. The entire system has been developed using Java and SWI-Prolog that runs on both Linux and Windows

A Morphological Analyzer to Enable English to Sinhala Machine Translation

Morphological analysis plays a key role in effective functioning of a parser of any machine translation system. Further, morphological analyzers are useful as supportive software tools to coin terms for a given language. This paper reports on the first morphological analysis system for Sinhala language. This comes out as a major step in the development of a Sinhala parser for machine translation from English to Sinhala. The paper describes how the morphological analyzer can detect grammatical information of a given Sinhala word, and generation of all possible forms of the given word. It is also presented how a language specialist can use our system to device Sinhala terms that are agreeable with Sinhala grammar. The system has been developed with the use of SWI-Prolog, and to runs on Windows and Linux.

Developing lexicon databases for English to Sinhala machine translation

Machine translation has been identified as one of the most challenging areas in natural language processing. In such systems, lexicon databases are fundamental to handle all translation requirements including lexicon, syntax and semantics information. This paper reports on the design and development of a lexicon database sub-system for English to Sinhala machine translation system. This system has customized the famous WordNet for English lexicon database, and structured three Sinhala dictionaries as per Japanese EDR system. SWI-Prolog has been used for the development of the core of the translation system.

Transliteration system for English to Sinhala machine translation

Machine translation is a challenging task in natural language processing. Out-of-vocabulary, handling proper nouns and technical terms are some major issues which are common to all machine translation systems. This paper presents a transliteration approach to machine translation from English to Sinhala. We have used finite state automaton to develop transducers for English to Sinhala transliteration. This approach can transliterate the text in original English and Sinhala words that are written using English letters. The transliteration system has been developed using SWI-PROLOG and prolog server page (PSP). English WorldNet and Sinhala Chatbot are used to test the transducers and reasonable results were achieved.

Optimization of multi-layer artificial neural networks using delta values of hidden layers

The number of hidden layers is crucial in multilayer artificial neural networks. In general, generalization power of the solution can be improved by increasing the number of layers. This paper presents a new method to determine the optimal architecture by using a pruning technique. The unimportant neurons are identified by using the delta values of hidden layers. The modified network contains fewer numbers of neurons in network and shows better generalization. Moreover, it has improved the speed relative to the back propagation training. The experiments have been done with number of test problems to verify the effectiveness of new approach.

Tacit knowledge modeling in Intelligent Hybrid systems

Knowledge modelling gives the intention of knowledge engineering which applicable for managing information systems. Tacit knowledge is the key issue of knowledge modelling aspect because all knowledge is rooted in tacit knowledge. This paper presents a research, which is incorporated of modelling of tacit knowledge. Here we have used an Intelligent Hybrid system for developing an approach for modelling tacit knowledge. The Intelligent Hybrid system is involved with artificial intelligent techniques, namely fuzzy logic and expert system technology. We primarily used fuzzy logic together with statistical technique of principle component analysis for modelling tacit domains. Tacit knowledge in Ayurvedic sub-domain of individual classification has been acquired through a questionnaire and analysed to identify the dependencies, which lead to make tacit knowledge in the particular domain. In the first place analysis was done using statistical techniques of principle components and the results were not compatible with the experiences of Ayurvedic experts. As such, fuzzy logic has been used to further model the Ayurvedic sub-domain. The result of the modelling of Ayurvedic domain using fuzzy logic has been compatible with the experiences of the Ayurvedic experts. It has shown 77% accuracy in using the tacit knowledge for reasoning in the relevant domain. The development has been done using Visual basic, FLEX expert system shell and the system runs on Windows platform. The Intelligent Hybrid system has been successfully applied for several tacit domains. Performances were very close to handling tacit knowledge by the human expert in tacit domain.

Theoretical based approach to English to Sinhala machine translation

Machine translation is a log felt need of the countries those who use English as a second language. Due to inherent complexity of natural languages, most of machine translation systems adopt rather ad-hoc strategies such as word level translation without concerning a proper theoretical basis. This has been a major reason for why the developments in natural language processing could not achieve as expected. This paper presents a theoretical-based approach to English-Sinhala machine translation through the concept of Varanagema (conjugation) in Sinhala Language. The theory of Varanagema in Sinhala language handles major language primitives including noun, verbs and prepositions. The concept of Varanagema also drastically reduces the number of word forms to be stored in the dictionaries of the machine translation system. The design, implementation and results of test versions of English-Sinhala machine translation system has been presented in the paper.

An approach to the development of commonsense knowledge modeling systems for disaster management

Knowledge is the fundamental resource that allows us to function intelligently. Similarly, organizations typically use different types of knowledge to enhance their performance. Commonsense knowledge that is not well formalized modeling is the key to disaster management in the process of information gathering into a formalized way. Modeling commonsense knowledge is crucial for classifying and presenting of unstructured knowledge. This paper suggests an approach to achieving this objective, by proposing a three-phase knowledge modeling approach. At the initial stage commonsense knowledge is converted into a questionnaire. Removing dependencies among the questions are modeled using principal component analysis. Classification of the knowledge is processed through fuzzy logic module, which is constructed on the basis of principal components. Further explanations for classified knowledge are derived by expert system technology. We have implemented the system using FLEX expert system shell, SPSS, XML, and VB. This paper describes one such approach using classification of human constituents in Ayurvedic medicine. Evaluation of the system has shown 77% accuracy.

Web-based English-Sinhala translator in action

Machine translation has been a potential solution for addressing the language barrier. In line with this, we have developed English to Sinhala machine translation system that can be accessed through the Internet. The translation system runs on a Web sever and can be accessed by an ordinary Web client. The core of the translation system contains seven modules, namely, English Morphological analyzer, English parser, Translator, Sinhala Morphological generator, Sinhala parser, Transliteration module and three Lexicon Databases. This core system has already been tested and used on standalone machines. The current project has extended the core system with the use of Prolog Server Pages to provide online access thereby opening the service to a wider audience.

On Computing Learning Ontology with Mental Attributes

Our research paper argues that learning can be made to be effective when it is learner-centered. Today learning is open to everyone. Knowledge acquisition is not restricted to vertical sources such as schools and colleges. Educational view point is to ensure that the learner has acquired knowledge through any means to ensure required levels of competencies. A learner may have acquired some knowledge without developing any analytical skills, effort to commit to learning or the ability to maintain a continuous process of learning. More seriously, the learner may not have developed adequate enthusiasm about a subject though it appears as if the learner has learned. From the educational viewpoint these are crucial issues that need to be addressed. As such measuring the development of mental attributes associated to a learning process is of great importance. Student learning styles vary enormously. Not all students respond positively to the same teaching techniques. Used well, learning technology can help turn the learning environment into a highly interactive, stimulating world where the student is totally embedded in the learning process. Learning technology can offer students control over their learning and flexibility, so that they can learn in the style that is most effective for each individual