Nowadays, social activity is tending to connect across nature, culture and language, translation’s development towards to the direction of further and wider, in other words, translation is a process, base on accuracy and smooth, from a sort of information in one language to other language. Means that convert a sort of information from strange to familiar. With the development of Information Technology and Computer Science, using computer to calculate and solve problem like translate to save human resource and relieve cost, which is imperative under the circumstances. Machine translation, also known as automatic translation, use computer translate a nature language another, generally means translate in sentence or passage of nature language. A branch of Nature Language Processing, relate to Computational Linguistics and Nature Language Understanding. Machine Translation Evaluation is essential to Machine Translation. We present a web service of diagnostic evaluation platform base on automatically constructed check-point. A check-point is a linguistically motivated unit(e.g. an ambiguous word, a noun phrase, a verb-obj collocation, a prepositional phrase etc.), which are pre-defined in a linguistic taxonomy. We present a method that automatically extracts check-point from parallel sentences. By means of check-point, our method can monitor a MT system in translating important linguistic phenomena to provide diagnostic evaluation. Our platform show in webpage, which offer help to MT developer and normal user make through a simple and friendly interface and operate. The evaluation result can be a reference of MT develop work, which not only a totally score, but also show ability of each check-point in different language phenomenas. Key words:Machine Translation;Diagnostic Evaluation of Machine Translation System;check-point;MVC structure
Some result picturem, below: