Smart vs. Solid Solutions in Computational Linguistics—Machine Translation or Information Retrieval

The 2018 Conference on Computational Linguistics and Speech Processing ROCLING 2018, pp. 256-265 ©The Association for Computational Linguistics and Chinese Language Processing Smart vs. Solid Solutions in Computational Linguistics －Machine Translation or Information Retrieval Su-Mei Shiue Department of Computer Science and Information Engineering National Taipei University of Technology t102599005@ntut.edu.tw Lang-Jyi Huang Department of Computer Science and Information Engineering National Taipei University of Technology t8599003@ntut.edu.tw Wei-Ho Tsai Department of Electronic Engineering National Taipei University of Technology whtsai@ntut.edu.tw Yen-Lin Chen Department of Computer Science and Information Engineering National Taipei University of Technology ylchen@csie.ntut.edu.tw Abstract Smart solutions and solid solutions are two different approaches for computational linguistics. To receive a satisfactory result at a minimum cost, smart solutions use special properties of the on-hand application. Solid solutions, in contrast, more generally analyze the application in order that these solutions can be used for a whole class of applications. By extending a well-maintained solid solution, it may be re-used once and again, thus saving time, money, and other resources in the long run. Therefore, the research focuses on solid solutions. It To implement smart solutions, machine translation is used while information retrieval is for solid solutions. Keywords: Smart Solutions, Solid Solutions, Computational Linguistics, Grammatical 256 Components, Natural Language. 1. Introduction 1.1 Destination of Computational Linguistics - Natural Transmission of Information The goal of computational linguistics is to reproduce the natural transmission of information by modeling the speaker’s production and the hearer’s interpretation on a suitable type of computing machine. This amounts to the construction of autonomous cognitive machines (robots) which can freely communicate in natural language. And the development of speaking robots is a real scientific task [1]. 1.2 Turing Test - Modeling the Mechanism of Natural Communication The task of modeling the mechanism of natural communication on the computer was described in 1950 by [4] in the form of an ‘imitation game’ known today as the Turing test The computer passes the Turing test if the man or the woman replaced by the computer is simulated so well that the guesses of the human interrogator are just as often right and wrong as with the earlier natural partner. In this way Turing wanted to replace the question “Can machines think?” with the question “Are there imaginable digital computers which would do well in the imitation game?” [1, 10] 1.3 Eliza Program – A Prototype of Smart Solution The Eliza program [2] simulates a psychiatrist encouraging the human interrogator, now in the role of a patient, to talk more and more about him- or herself. Eliza works with a list of words. Whenever one of these words is typed by the human interrogator/patient, Eliza inserts it into one of several prefabricated sentence templates. For example, when the word mother is used by the human, Eliza uses the template Tell me more about your ___ to generate the sentence Tell me more about your mother. Eliza program is the prototype of a smart solution in that it exploits the restrictions of a highly specialized application …

Content truncated to 500 words