Whilst semantic relations apply to components of a synset since are all mutually synonyms, words can also be connected to other words through lexical relations, including antonyms (opposites of each other) and derivationally related, as well. 20 Neural networks have broad applicability to real world where internet source tends to be the target to gather information especially on cross translation strategies. In fact, they have already been successfully applied in many industries as sales forecasting, industrial process control, customer research, data validation, risk management, target marketing, etc.
Neural networks are best at identifying patterns or trends in data that are well suited for prediction or forecasting needs including: specific cross translations, recovery of telecommunications from faulty software; interpretation of multi-meaning Chinese words; under sea mine detection; texture analysis; three-dimensional object recognition; hand- written word recognition; and facial recognition. WordNet when in tandem with Artificial Intelligence implemented through Neural Network would prove to be a very promising system, add to that, such system in a web-based environment.
A web-based software (www. gliukaz. co. uk/dictionary) was developed to create cross-language translation system using English as its base language for cross-language translations (i. e. Arabic ? English ? Lithuanian). Cross-language translation was made as accurate and as straightforward as possible. The abovementioned software was made utilizing MS Visual Studio 2008(C# and ASP . NET) and 3. 0 Framework, hand in hand. Appendix D shows the source code designed for this particular software.
Base dictionary database implemented together with WordNet database using with MS SQL 2005 enterprise server. WordNet database used to for getting English language definitions. (As seen in figure 5 and 6) Translation accuracy was achieved by splitting words in to meanings based on English. Figure 7 thru 9 shows some examples of word translations done from Lithuanian to English, English to Esperanto, and Lithuanian to Esperanto, respectively.
Successful implementation of Cross Language translation in a web-based environment through Artificial Intelligence in the realm of neural networks had been done with the aid of WordNet® (www. gliukaz. co. uk/dictionary). Accuracy has been assured by using English as base language which meant translating both input and output word languages to English (i. e. Arabic ? English ? Lithuanian). The software was made possible by using both MS Visual Studio 2008(C# and ASP . NET) and 3. 0 Framework during development. Further studies however can be seen surrounding this chosen topic.
The author recommends implementing and allowing user to add words and using user rating system as good supervised learning example. Another idea is to create a server side application that would analyze WordNet® data and create new data base, a new network similar to WordNet® covering more than just English but more languages that are awaiting to be implemented with such system as WordNet®. The author further recommends further development of this project into accepting inputs with lengths such as phrases as well as sentences.
Beckwith, R. , Miller, G. , Tengi, R. (Unknown). Design and Implementation of the WordNet Lexical Database and Searching Software. 62-79. Chung, W. , Wong K. , Soon, P. (2007) Journal of Manufacturing Technology Management: An ANN- based DSS system for quality assurance in production network. 836-857. Diekema, A. R. (2003). Translation Events In Cross-Language Information Retrieval: Lexical Ambiguity, Lexical Holes, Vocabulary Mismatch, And Correct Translations. 1-21. Dietterich, T. G. (Unknown). Machine Learning. 1-14.