lately , Tsinghua University in GitHub Open source an artifact called the king of ten thousand words （Want Wrong）, It claims to be the first open source online reverse dictionary supporting Chinese and cross language query .
What is a reverse dictionary ? An ordinary dictionary tells you the definition of a word , The reverse dictionary is just the opposite , It can tell you which words match the meaning of the description you enter . The following is a screenshot of wanciwang online reverse dictionary , It demonstrates an example of reverse word lookup , input “ The mountain is very high ”, The system will return a series of model representations “ The mountain is very high ” Meaning words , for example “ Gao Jun ”,“ Towering ” etc .
What can reverse dictionaries be used for ?
* solve “ Tongue tip phenomenon ”（tip-of-the-tongue, It is also called that words can't be spoken ）, That is, the problem of temporarily forgetting words
* Help language learners learn , Consolidate vocabulary
* Improve the quality of life of patients with word selection aphasia , The symptom of the disease is that an object can be identified and described , But I can't remember the name of the object
In short , Even today's young people will be poor in words , When you are poor in words , This tool can come in handy , Like when you're flustered and can't make up your mind , I don't know which word to use , You can enter it , Will be hesitant , Don't feel flustered, etc 100 Words . And the part of speech and rhyme can also be customized .
besides , It also supports Chinese , English , Chinese-English , There are many different ways of transformation, such as English and Chinese .
The overall system architecture is also very simple ：
Wan ciwang's core model was the natural language processing laboratory of the computer department of Tsinghua University （THUNLP） Published in AAAI-20 A multi-channel reverse dictionary model proposed in a paper on ：Multi-channel
Reverse Dictionary Model [ paper ] [ code ], The model architecture is shown below ：
About natural language processing and social humanities Computing Laboratory of Tsinghua University
Laboratory of natural language processing and social humanities computing, Department of computer science, Tsinghua University （THUNLP） Founded in 20 century 70 Late S , At first, he was engaged in the research of Chinese information processing under the leadership of Professor Huang Changning , It is the earliest research on natural language processing in China , Influential scientific research institutions , It is also the Chinese information society （ National first level society ） Affiliated unit of the professional committee of Computational Linguistics . The academic leader of the laboratory is Professor Sun Maosong , The team of laboratory teachers also includes Professor Liu Yang and Associate Professor Liu Zhiyuan . The laboratory carries out systematic and in-depth research on the cutting-edge basic topics of natural language processing with Chinese as the core , The research field covers the core issues of computational linguistics, social computing and humanistic computing , In recent years 973,863, With the support of National Natural Science Foundation of China and other projects , Where are the teachers and students in the laboratory IJCAI,AAAI,ACL,EMNLP And other international top conferences and journals , And CMU,NUS,Google And other international famous schools and enterprises have long-term good cooperative relations , Most of the outstanding graduates trained go to Tsinghua University , Google , Baidu , Ali , Microsoft and other famous universities and enterprises .
WantWords from THUNLP Development and maintenance , The project instructors are Professor Sun Maosong and Associate Professor Liu Zhiyuan , The development team members include fan Chao , Lei Zhang , Yang Yanhui .
at present ,Wantwords Already in GitHub Superscript star 1K, Cumulative branch 56 individual
Technology