A Probe into the New Mode of TCSL from the Perspective of the Fourth Paradigm

【摘要】研究對象與理論成果的主客不分離是研究人文社會科學與自然科學的最鮮明的特征，它們都是彼此交織的。對身處學術世界中的人文社會的學術態度決定了相關研究的范式取向。對這研究對象從生成、積累到發展壯大的歷時過程的表征與映射，積聚了研究對象的大數據集合。

【關鍵詞】第四范式;對外漢語;數據庫

【Abstract】One of the most distinct and essential characteristics of humanities and social science research is that it is not subject-object separation from its research object， but intertwined with each other. Then， how to look at the human society in which the world determines the paradigm orientation of relevant research. The big data set accumulated by the research object is the representation and mapping of the diachronic process from generation， accumulation to development and growth of the research object.

【Key word】the Fourth Paradigm; TCSL; Corpus Aided

【作者簡介】苑洋（1988-），女，河南商丘人，三亞學院人文與傳播學院，助教，碩士，研究方向：漢語國際教育。

【基金項目】本文系海南省哲學社會科學規劃課題《基于第四研究范式的漢語教學新模式建構》（HNSK（GJ）19-06）階段性成果。

Nowadays， the love to the “algorithm” has spread to all fields of academic and science， and furthermore， the proposal of “fourth paradigm” has accelerated the rapid development of data in these fields. Meanwhile， linguists field has also begun to enter the era of data.

I. Concept of “Fourth Paradigm”

In January 2001， Dr. Jim Gray， a computer scientist at Microsoft Research Institute， put forward the “fourth paradigm” of scientific research， which is data-intensive scientific research paradigm. The purpose of this paper is to call on computer scientists to devote themselves to the development of effective tools to serve the whole research process of “data acquisition， data classification management， data analysis to data visualization”.The research paradigm shift in big data era has been widely concerned since Initiative on Research and Development of Big Data published in 2012. Big data has a profound influence in various fields because of many types of data structure， fast speed， wide channels and far-reaching influence in various fields. In the field of education， learning analysis and educational data mining have been developed rapidly， and at the same time， it has also caused the transformation of research paradigm. The so-called paradigm can be understood as a scaffolding in the field of scientific research， which is a research method that can be followed， including the object， process， results and other elements of the research. The traditional research paradigm is lack of evidence-based research， and the scientific nature is relatively low. The emergence of big data provides a new research method for scientific research-data-intensive science， which is to use IT technology to obtain， process and analyze data. This data-based research method is called the fourth paradigm.With a little analysis of some frontier topics or research hotspots in the field of humanities and social sciences， we can see that the new paradigm of data-intensive humanities and social sciences research has also begun to emerge. Taking language research as an example， the research model based on corpus has become a broader consensus and general scientific research practice in the world. The data types of corpus are expanding day by day， written corpus， oral corpus， text corpus， phonetic corpus and the continuous construction of all kinds of language knowledge base， and so on. At the same time， the content of language research with the help of corpus is expanding， which involves almost every aspect of traditional language research， and includes some frontier fields.

II. the “Fourth Paradigm” and the New Model of TCSL

With the rapid growth of data and the development of computer algorithms， computers will not only be able to simulate and simulate， but also learn， sum up， analyze， reason， sum up， and get theories， that is to say， in the past， some of the work done by scientists such as Newton， Einstein， by Saussure， Chomsky and other linguists can be done by computers， which opens up a broad prospect for the study of language science. Such research， we call it the “e-study” of language. For the e-study of language， our assumption is that， through the way of big data， on the basis of large data using language material， using the Internet search language study is used as the study tips and reference， to compare and confirm each other， make a sublation， used as a study group activity parameters， to help the research conclusions. Big data thinking： the fourth paradigm of the new concept of language research is a data-intensive research paradigm， the essence of which is to “let the data speak” in the research.

1. Corpus Aided Teaching of TCSL

Corpus is a kind of resource database， which is based on computer and real language materials， and is processed （analyzed and processed）. The use of language materials in corpus is similar to the use of language itself， which is representative and authentic. The use of language materials in corpus is similar to the use of language itself， which is representative and authentic. The introduction of corpus into second language teaching and learning provides a new possibility for second language teaching and learning. Corpus-driven learning model is a new language learning model in recent years. Since Tim Johns put forward a foreign language learning method based on corpus data-driven learning model in the early 1990s， corpus has been inextricably related to language teaching.This language teaching method guides students to use index software to query the corpus. These corpus come from the spoken or written language of the native speakers. By querying the real corpus in the use of the language， rather than the examples designed by the textbook compilers， the learners can more truly understand the use of the target language in practical use. At present， the influential corpus is： CCL Corpus of Peking University， Modern Chinese Corpus of National Language Committee， BCC Corpus of Beijing Language University， Dynamic Composition Corpus of HSK， DCC Dynamic Circulation Corpus of National Language Resource Monitoring and Research Planar Media Language Center and so on. However， under the influence of the purpose of database construction and the type of end-user， most of the corpus is often only a small group of corpus researchers engaged in language research tools， only a small number of corpus is directly used in language teaching. These corpus contain rich data， but the corpus without fine particle tagging and classification is of limited help to learners.

It is found that learners are more likely to achieve good language learning results when they pay attention to and deal with enough real language examples. Therefore， under the guidance of corpus-driven learning model for autonomous learning， real language input， self-exploration and discovery， learners can acquire language knowledge more effectively. Although the corpus-driven model is helpful to TCSL， at present， our country can be directly used in TCSL， there are very few high-quality corpus that can be shared， and the content of large-scale and general corpus can not be directly connected with TCSL. In order to better apply corpus to TCSL， the corpus and application platform for TCSL need to be developed urgently.

2. Research and Development of APP in TCSL

The serious shortage of international Chinese teachers is the biggest problem in TCSL at present. The universal mobile Internet and the increasingly mature mobile phone App can provide very convenient Chinese learning conditions for foreign Chinese learners. According to statistics， in the current App application market， the number of educational App has exceeded hundreds of thousands， the content basically covers most disciplines of basic education， but there is only a lack of App applications related to TCSL. Mobile phone mobile App can break the shackles of “time” and “space” in TCSL. Nowadays， foreign Chinese learners are distributed all over the world. The huge geographical leapfrogging and the serious shortage of international Chinese teachers make these students unable to obtain correct and systematic Chinese learning， and the emergence of mobile App can solve this problem very well. With the emergence of mobile phone mobile App， foreign Chinese learners can learn Chinese systematically anywhere in the world. This new Chinese learning method is very convenient. All they have to do is turn on their cell phone， App. In addition， this new way of learning is to break the traditional way of learning. In this new way of learning， people can learn Chinese systematically at any time without the limitation of time， and can choose the length of learning time according to their own situation. In addition， this new learning method can enable foreign Chinese learners to make full use of trivial time to master a relatively complete Chinese knowledge block， just as in the face of a bunch of seemingly disorganized knowledge fragments， after one piece of daily intake and accumulation， a complete knowledge jigsaw puzzle will eventually be formed. This unique advantage is beyond the reach of the traditional way of learning. In addition， mobile App is rich in content， which helps to enhance students interest and enthusiasm in learning. Adding pictures， audio and video to the mobile APP will greatly reduce the obstacles to TCSL， at the same time， it can also cultivate students ability of abstract thinking， make the original obscure knowledge more flexible， improve students interest in learning， and promote students absorption of teaching content.

Nowadays， the big data era has brought us a lot of information， the amount of information is huge， and the structure of the complex big data often implies peoples unexpected findings. In the era of big data， the establishment of a diversified corpus suitable for learners and the combination of it with the application of network interactive technology can change the passive situation of language teaching by relying solely on teaching materials， solve the problems of authenticity and diversification of language learning materials， adapt to the actual needs of three-dimensional， networked， individualized language teaching and learning， and make the TCSL towards individualization. Regardless of time and place restrictions， active learning direction development.

III. Epilogue

One of the most distinct and essential characteristics of humanities and social science research is that it is not subject-object separation from its research object， but intertwined with each other. Then， how to look at the human society in which the world determines the paradigm orientation of relevant research. The big data set accumulated by the research object is the representation and mapping of the diachronic process from generation， accumulation to development and growth of the research object. As a scientific research object in the form of data， it is not only the representation of the real world represented by data， but also the medium of knowledge discovery based on data. Big data contains a feature of knowledge discovery： “findability”. This is something that language researchers can master after learning. In big datas time， this is a tool for language researchers， which is worth taking the time to master. In the era of big data， driven by the fourth paradigm， linguistic research depends not only on the level and proficiency of researchers using databases， but also on the maturity of research ideas and the spirit of serious search.

參考文獻：

[1]徐盛桓.第四范式.語言研究的新理念[J].英語研究，2016（10）.

[2]黃芙菊，孟德騰.語料庫數據驅動模式在對外漢語教學中的應用[J].中國教育信息化，2013（12）.

[3]魏向清.語言研究“第四范式”之思[J].外語研究，2014（08）.