摘要
在当今信息时代,人机口语对话系统有着广泛的应用需求。口语语言中包含有大量的省略、停顿、重复、自我纠错以及不合语法的现象等,因此在汉语人机口语对话中,口语语言的分析理解是人机对话系统实现的关键和难点。现在的口语对话系统语言的分析处理方法大多采用模板匹配处理方法,然而口语语言的灵活性使得模板数量过于庞大,而且导致系统准确性不高。本文着重研究汉语口语对话系统中的口语语言分析,尝试用内涵概念分析的思想,把对语言的分析上升到概念层次上进行分析来解决该问题。本文研究工作背景来源于上海市科委科技项目限定领域口语对话系统——交通领域口语对话系统SHJTQ, SHJTQ提供上海市任意两个地点之间、在不同的交通方式(步行、自行车、出租车、公交车)下的交通路线信息的查询。
目前限定领域的口语语言分析理解的方法大致可以分为两类:概率统计方法和规则分析方法。概率统计分析技术主要基于语言结构的统计特性,缺乏智能性和可靠性。规则分析方法又分为逻辑分析方法和概念分析方法两类。逻辑分析方法以蒙太古的语义学为代表,用模型论来表示片断英语的语义,但要处理真实文本、全面地解释汉语语义,仍然感到逻辑分析的局限性。概念分析是后继的逻辑学家维特根斯坦、奥斯汀、塞尔这些哲学家提出的,语言哲学家以及心理哲学家关心的是有关心理、感觉、情感一类词汇的概念分析,但不注重对于指称实体的词类及其概念的研究。
现在的口语对话系统都是在应用层面进行语言的分析处理,提出的解决方法大多采用字串匹配的方法或是在字串匹配的基础上加入一些处理方法。这种分析方法最大的缺点就是字串不同或字串次序变化,分析就会失败。因此无法解释灵活多变的口语语言。本文提出了内涵概念分析的思想,把对语言的分析上升到概念层次上进行分析,口语语言虽然灵活多变,但表达的概念是一样的,从而解决了模板匹配解决不了的问题。
从实现角度讲,除音调外一字符串(如短语、句)的语音要用1K数据存储空间,对通常对话的语音处理将占用超大量存储空间。如果改为一汉字一模板,两千常用汉字的语音信息共2K*1K数据,利用汉语是字组合表达概念直接耦合的优点,将字语音模板作为单位,字组合即为语音模板信息组合,可大大减少语音数据,为语音对话使用开创了可行前景。可不限用户使用规定的表达格式、模板可自由表达。但由此带来语言处理的复杂性和重要性。本文充分利用汉语概念内涵模型思想,实现了专用领域内对话词语的概念分析,并获成功。
本文研究了SHJTQ中词汇(主要是交通工具类词汇)的内涵特征,提出了名词具有“定义特征”和“情景区分特征”两个概念。在不同语境下,词汇的凸现特征(情景区分特征)有所不同。提出用一种“E—A-V”(实体-属性-值)的方法表征名词的概念。本文研究了SHJTQ中用户问句,针对用户问句多为疑问句的特点,借助言语行为理论的思想,对SHJTQ系统用户查询问句进行了言语行为分类。分析研究了SHJTQ口语语句的内涵概念,根据用户查询问题的分类,逐个将各类用户查询例句了进行概念分析,解决了字串匹配不能解决的口语语言中的各种变异现象,为汉语口语语言的理解研究提供新的思路。
本文介绍了限定领域对话系统口语语言的概念分析方法在SHJTQ中的具体应用。着重分析了SHJTQ语言分析模块的设计等,同时本文给出了系统的测试结果和分析。
本文研究的创新点主要在如下几方面:
1.用概念分析的方法分析汉语口语语言,有别于传统的应用层面上的字串匹配分析方法。从概念层面分析解释了SHJTQ中口语语言,解释了口语语言在形式上灵活多变,但表达的是同一概念。另外采用概念分析方法,汉语和其他语言(如英语等)在体态(形态、时态等)方面的差别就会退化,有助于实现多语种的口语对话。第三,在具体实现方面,有了口语语言的概念分析,语音识别需要的模板量就可以大大减少,可以推进口语对话系统的发展。
2.采用“E—A-V”(实体-属性-值)的表示概念语义模型,表征了名词的多义性。本文采用的是陆汝占先生的内涵逻辑分析即概念分析的基本思想,即对一个词语所表示的概念进行分解,求解出上位概念、下位区分概念、定义属性特征以及扩展特征;解释了词语、指称实体、概念三者关系。认为名词是指称实体的词语项,实体包括物理实体和抽象实体两类。名词具有内涵性质即语义特征,提出了“定义特征”和“情景区分特征”两个概念。分析了SHJTQ中词汇(主要是交通工具类名词词汇)的内涵特征,将内涵特征理论引入到汉语语言研究,通过内涵特征来解释纷纭复杂的汉语语义是一个新尝试。
3.用内涵概念分析的方法研究了SHJTQ口语语句,根据用户查询问题的言语行为分类,逐个将各类用户查询例句进行概念分析,分析了简单完备用户表述语句、带变异的表述、不完备表述语句,把用户不规范的表达转变成概念层次上规范的查询表达,从而解决了字串匹配无法解决的口语语言灵活表达的问题。实现了概念分析指导下的限定领域口语对话系统。经测试,系统准确性较高。
Nowadays, man-machine spoken dialogue system (SDS) is an active research field with wide application demand. There exist lots of non-written languge phenomena in spoken language, such as ellipsis, pause, parenthesis, repeat and re-start. Most of spoken language sentences are grammatically incorrect or ill-formed. So in Chinese SDS, the key and difficult problem is how to understand spoken language. Template matching processing is a popular method to do this in currunt Chinese SDSs. But the flexibility of spoken language makes the template amount very huge, so it influences system’s accuracy. In this thesis, we focus on understanding spoken language in Chinese SDS, attempts to analyze the spoken language phenomena in conceptual level. The background of this research work is a Chinese SDS――SHJTQ (shanghai Jiaotong Query System), which can provide information about the best route between any two sites in Shanghai.
There are two methods for handling spoken language: statistic method and rule-based analytical method. Statistics analysis does it mainly according to the statistic characteristics of the language structure, regardless of semantic features. So this method lacks of intelligence and reliability. Rule-based method can be divided into two types: logic analysis and concept analysis. Montague semantic is the representative method of logic analysis method. Using model theory, it successfully gives the meaning of fragment English. But it fails in handling true text, especially in explaining Chinese. Concept analysis is put forward by philosophers such as Wittgenstein, Austin and Searle. What the language philosophers and mental philosophers concern is analyzing concept of the word in mental state, felling, emotion. But they seldom pay attention to analyze the concept of the word of reference entity.
Most of currunt SDSs analyze spoken language in the applied level, string matching method or combined with some other processing methods is adopted. The severe disadvantage of these kinds of methods is when the string changes or the order of string changes, the analysis will fail. So it can not deal with the flexibility of spoken language. In this thesis, we put forward the thought of intension concept analysis, and analyze the spoken language in the upper concept level. So we explained why such different character strings (expression type) express the same concept.
From the aspect of realizing, to store the pronunciation of a string (for example phrase, sentence), we needs 1K data storage space (ignored tonality information). That is to say, ultra massive storage space is required to process the pronunciation of usual dialog. If changes the method, a Chinese character correspond to a template, the pronunciation information of 2000 common character altogether will take limited 2K*1K space. This is due to Chinese has the characteristic that the character combination express concept. So this method can reduce the pronunciation data greatly. But it brings a new question, that is, the language processing become more complex and important. In thhis thesis, using the thought of Chinese intension concept model, we realized conceptual analysis of words in domain-specific SDS, and attained successfully.
The intention characteristic of the word in SHJTQ domain (mainly the word belongs to vehicles) is analyzed in this thesis. We purposed a noun has 2 concept characteristic: "definition characteristic" and "the situation distinction characteristic". The appearance characteristic of a word (the situation distinction characteristic) varies in different situation. We proposed a kind of E-A-V (entity-attribute-value) method to represented noun’s concept. In our domain-specific SDS– SHJTQ, most of user’s spoken language are interrogative sentences. After combining the thought of speech act theory, we reclassified the users’query sentences in SHJTQ, and this directly helped our spoken language analysis. We analyzed the intention concept of spoken sentences in SHJTQ. According to the classified user’s query questions, we analyzed the truly user’s query sentences in the upper concept level one by one. And we also analyzed several variant phenomena in spoken language. It’s a new way of thinking to understand Chinese spoken language.
Concept analysis of spoken language in domain-specific spoken dialogue system is applied into SHJTQ. We propose the whole design of the system. Sub-modules of the language-understanding module, such as POS tagging, robust parsing and concept analysis are especially discribed. System performance is tested and analyzed.
The novelties of this paper reside in the following aspects.
1. Proposed a method called conceptual analysis to analyze Chinese spoken language, which is different with traditional string match analysis method in the application level. Analyzed spoken language in SHJTQ from the aspect of conceptual analysis, explained the question that spoken language is formally flexible, but expresses the same meaning. The other advantage of concept analysis is it helps to establish multi-language SDS. For example, Chinese language has no tense, single plural variety, but other language (like English) contains the variety of the appearance, tense...etc., so if we analyze it in the upper concept level, these appearance phenomena will become weak. The third, in realizing SDS, the template amount which the speech recognition is required has been possible to reduce greatly by using this method. It might give an impulse to the development of spoken language dialog system.
2. Proposed using the E-A-V (Entity– Attribute– Value) model to represent polysemy of noun modification. We adopted the thought of intension logic analysis (put forward by Prof. Lu Ru-zhan), decomposed the concept that a word expressed into upper concept, lower concept, defined attribute and expanded features. So, we can explain the relation of word, denotation entity and concept. Our research work also indicates a noun is a phrase item of a denotation entity, it owns two concept characteristic: "definition characteristic" and "the situation distinction characteristic". Next, we analyzed the intention concept of spoken sentences in SHJTQ. It’s a new way for understanding Chinese spoken language.
3. By using intensional concept method, we analyzed the truly user’s query sentences in the upper concept level. And we also analyzed several variant phenomena in spoken language. Our solution for general language phenomena such as ellipsis, reference and negation is also described. The system performance is tested and analyzed.
引文
[1] Allen James. Natural Language Understanding. The Benjamin/Cummings Publishing Company, Inc. 1995.
[2] Allen J., Schubert L., Ferguson G., Heeman P.,Hwang C., Kato T., Light M., Martin N., Miller B., Poesio M., Traum D. The TRAINS Project: A case study in building a conversational planning agent. Journal of Experimental and Theoretical AI, 7:7-48, 1995.
[3] Aust H, Oerder M, Seide F. The Philips automatic train timetable information system. Speech Communication, 17:249-262, 1995.
[4] Sanderman, J. Sturm, E. den Os, L. Roves, and A.Cremers. Evaluation of the Dutch Train Timetable Information System developed in the ARISE Project, In: Proc. of IVTTA, pp91-96, 1998.
[5] Barwise Jon, and John Perry. Situation and Attitudes. Cambridge, MA: Mit press. 1983.
[6] Barwise, Jon. "Scenes and Other Situations", The Journal of Philosophy, vol. 77, pp. 369-97. 1981.
[7] Barwise, Jon and John Perry. "Semantic Innocence and Uncompromising Situations", in P. French, T. Uehling, and H. Wettstein, eds, Midwest Studies in the Philosophy of Language, vol. VI (Minneapolis: University of Minnesota Press). 1981a.
[8] Barwise, Jon and John Perry. "Situations and Attitudes", Journal of Philosophy, vol. 77: 668–691. 1981b
[9] Barwise, Jon, and John Etchemendy. The Liar: An Essay on Truth and Circularity. New York: Oxford University Press. 1987.
[10] Barwise, Jon, Jean Mark Gawron, Gordon Plotkin, and Syun Tutiya, eds., Situation Theory and Its Applications, Stanforcl: CSLI Publication. vol. 2, 1991.
[11] Barwise,Jon. "The Situation in Logic. CSLI Lecture" Notes Number 17.Stanforcl:CSLI Publication. 1989.
[12] B. Pellom, W. Ward, S. Pradhan, The CU Communicator: An Architecture for Dialogue Systems, In Proc. of ICSLP 2000, Beijing China, 2000.
[13] Chafe, W L. "Language and consciousness." Language.50.111-33. 1974
[14] Chafe, W L. "Givenness, contrastiveness, definiteness, subjects, topics and point of view." C. N. Li,ed...pp27-55. 1976
[15] Chao HUANG, Peng XU, Xin ZHANG, et al., LODESTAR: A Mandarin Spoken Dialogue System for Travel Information Retrieval, in Eurospeech, pp. 1159-1162. 1999.
[16] Chomsky, N. "Essays on Form and Interpretation" , North-Holland , New York . 1977.
[17] Chomsky, N. Syntactic Structure Hague:Monton.. 1957.
[18] Chomsky, N. Aspects of the Theory of Syntax Cambridge ,MA:MIT Press. 1965.
[19] Chomsky, N. and J.Katz. "What the Linguist is talking about" , Journal of Philosophy1974 .
[20] D. Jurafsky, C. Wooters, G. Tajchman, J. Segal, A.Stolcke, E. Fosler and N. Morgan, "The Berkeley Restaurant Project", Proc. ICSLP, pp2139-2142. 1994.
[21] E. den Os, L. Boves, L. Lamel, and P. Baggia, "Overview ofthe ARISE project", Proc. Eurospeech, pp1527-1530. 1999.
[22] Fillmore, C. "The Case for Case", in E.Bach and E.Harms eds. "Universals in Linguistic Theory", New York. 1968.
[23] G. Castagnieri, P. Baggia, and M. Danieli, "Field Trials of the Italian ARISE Train Timetable System", Proc. TVTTA, pp97-102, 1998.
[24] Helen Meng, et al. "ISIS:A Leanring System w ith Combined Interaction and Delegation Dialogs", in Eurspeech, pp1551-1554, 2001.
[25] H.P. Grice. "Logic and Conversation", Syntax and Semantics, Speech Acts, Vol.3, Academic Press, pp41-58, 1975.
[26] J. Cassell. "Embodied Conversation: Integrating Face and Gesture into Automatic Spoken Dialogue Systems", In Luperfoy (ed.), Spoken Dialogue Systems. Cambridge, MA: MIT Press. 1998.
[27] J. Glass, et al. "The MIT ATIS System: December 1994 Progress Report", Proc. DARPA Spoken Language Systems Technology Workshop, pp252一256. 1995.
[28] J. Glass, G. Flammia, D. Goodine, M. Phillips, J. Polifroni, S. Sakai,S. Seneff, and V Zue, "Multilingual Spoken-Language Understanding in the MIT Voyager System", Speech Communication, 17, pp1-18. 1995.
[29] J.Katz, J.A.Fodor. "The Structerre of a Semantic Theory", Language 39:170-210. 1963.
[30] J. Peckham, "A New Generation of Spoken Dialogue Systems: Results and Lessons from the SUNDIAL Project", Proc. Eurospeech, pp33-40. 1993.
[31] J. Lyoms. Semantics, Cambridge University Press, Cambridge. 1972.
[32] J. Lyons. Introduction to the theoretical Linguistics, Cambridge University Press, Berkeley. 1968.
[33] K. Papineni, S. Roukos, and R. Ward. "Maximum Likelihood and Discriminative Training of Direct Translation Models", Proc. ICASSP, pp189-192, 1998.
[34] Leech, G. N. Principles of pragmatics, London: Longman. 1983.
[35] M. Bates, R. Bobrow, P. Fung, R. Ingria, F.Kubala, J. Makhoul, L.Nguyen, R. Schwartz, and D Stallard. "The BBNIHARC Spoken Language Understanding System", Proc. ICASSP, pp111-114. 1993.
[36] M. Blomberg, R.Carlson, K. Elenius, B. Granstrom, J.Lindell, and L. Neovius, "An Experimental Dialogue System: Gustafson, Waxholm", Proc.Eurospeech ,pp1867-1870. 1993.
[37] MICHAEI F M. Spoken dialogue technology: enabling the conversational user interface. ACM Computing Surveys, 134(1):90-169. 2002.
[38] Morris, C. Foundations of the Theory of Signs, Chicago: University of Chicago Press. 1938
[39] Prager, etc. Question-answering by predictive annotation. Proceedings of SIGIR. 2000.
[40] R.Montague. "English as a Formal Language". In Formal Philosophy.R.Thomasom(ed.). Yale, New Haven., pp188-221. 1974a.
[41] R. Montague. "Universal Grammar". In Formal Philosophy.R.Thomasom(ed.). Yale, New Haven., pp222-246, 1974b
[42] R. Montague.“The Proper Treatment of Quantification in Ordinary English”. In Formal Philosophy. R. Thomasom(ed.). Yale, New Haven, pp247-270. 1974c.
[43] RU Haibo, CHEN Yining, LIU Jia, LIU Runsheng. A Dialogue System Based On Keyword Spotting Technology, In Proc. of the 2003 IEEE International Conference on RoboticsJntelligent Systems and Signal Processing, Changsha, China, pp38-42, 2003.
[44] Rudnicky, J.M. Lunati, and A. Franz, "Spoken Language Recognitionin an Ofice Management Domain", in Proc. of ICASSP, pp829-832, 1991.
[45] Schank, R.C, (ed). Conceptual Information Processing, Amsterdam:North-Holland. 1975
[46] Schank, R.C and Abelson, R. Scripts, plans, Goals and Understanding, Hillsdale, NJ:Lanrence Erlbaum. 1977
[47] Searle, J.R. "Austion on locutionary and illocutionary acts". Philosophical Review, vol.77, 405-424. 1968.
[48] Seneff, S. "TINA: A Natural Language System for Spoken Language Applications". Computational Linguistics, Vol 18, No. 1, pp 61-86, 1992.
[49] S. Marcus, et al. "Prompt Constrained Natural Language Evolving the Next Generation of Telephony Services", Proc. ICSLP, pp857-860. ,1996,
[50] S. Sene, E. Hurley,R. Lau, C. Pao, P. Schmid, and V. Zue, "Galaxy-II A Reference Architecture for Conversational System Development", in Proc. of ICSLP, pp931-934. 1998.
[51] Stolcke, A., Ries, K., Coccaro, N., Shriberg, E., Bates, R., Jurafsky, D., Taylor, P., Martin, R., Van Ess-Dykema, C., and Meeter, M.: Dialogue Act Modeling for Automatic Tagging and Recognition of Conversational Speech. Computational Linguistics, vol 26, no 3, 339–373. 2000.
[52] W. Ward, and S. rssar, "Recent Improvements in theCMU Spoken Language Understanding System", in Proc. of ARPA Human Language Technology Workshop, pp213-216, 1996.
[53] Y. Wilks. "An Arlifcial Intelligence Approach to Machine Tranglation", in Computer Modols of Thoughtal language,R.Shark and K.Colby(eds),W.H.Freeman Co. 1973.
[54] Y. Wilks. "A Preferential Pattern-Seeking Semantics for Natural Language Inference", Artificial Intelligence. York: Oxford University Press. 1975
[55] Zdravko Kacic. Advances in spoken dialogue system development, in proc. of IEEE ISIE’99 Bled, Slovenia. 1999.
[56]拜战胜,蓝岚,彭佳红,陈哲.对话系统中控制模型的比较研究.郑州大学学报(理学版). 38(4):112-116, 2006.
[57]蔡义发.自然语言理解的研究与发展.计算机应用与软件. 9(3):1-6, 1992.
[58]陈芳.“智能信息系统中计算机语言生成的研究”.国家自然科学基金课题技术报告. 1998.
[59]程雨民.格赖斯的“会话含义”与有关的讨论.国外语言学,第1期, pp19-25,1983.
[60]戴礼荣.人机语声对话特点及系统设计.安徽大学学报(自然科学版). 21(4):52-56, 1997.
[61]冯志伟.《自然语言的计算机处理》.上海外语教育出版社. 1996.
[62]郭荣.人机对话系统中的语言分析[硕士学位论文].上海交通大学. 2003.
[63]郭艳华,周昌乐.自然语言理解综述.杭州电子工业学院学报.第2期, pp58-64, 2000.
[64]何兆雄.《语用学概要》.上海外语教育出版社2000.
[65]黄曾阳.《HNC(概念层次网络)理论》.北京:清华大学出版社. 1998
[66]靳光瑾.《现代汉语动词语义计算理论》.北京:北京大学出版社. 2001.
[67]蒋绍愚.两次分类.中国语文.第5期, pp323-330, 1999.
[68]李德华,刘根辉.面向信息处理的语境形式化研究.中文信息学报. 18(3):32-38, 2004.
[69]李沐,姚天顺.一种基于优先关系的LSD分析算法.计算机研究与发展.第5期, pp597-602, 2001.
[70]刘开瑛,郭炳炎.自然语言处理.科学出版社. 1991
[71]陆汝占,靳光瑾.蒙太古语义学.语言文字应用, 16(4): 47-52, 1995.
[72]陆汝占,靳光瑾.汉语语义表示、解释、计算.汉语计算国际会议论文集. 2000.
[73]陆汝占,靳广瑾.“是”的内涵逻辑语义解释.全国第六届计算语言学联合学术会议论文集. 2001.
[74]陆汝占,高峰,陈玉泉,靳光瑾.汉语内涵逻辑及其应用.辉煌二十年——中国中文信息学会二十周年学术会议论文集.北京. pp89-95, 2001.
[75]戚雨村.语用学说略.外国语,第4期, 1988.
[76]邵敬敏.《现代汉语疑问句研究》.上海:华东师范大学出版社. pp73-83, 1996.
[77]宋春阳.面向信息处理的现代汉语“名+名”逻辑语义研究[博士学位论文].上海师范大学. 2004
[78]讚井唯允.语用上的具体化与一般化.大河内主编.日本近、现代汉语研究论文选.北京语言学院出版社. 1993.
[79]韦宏利,韦宏利,付蓉.公交线路智能查询系统的设计.现代电子技术.第12期, pp16-17, 2003.
[80]翁富良,王野翊.《计算语言学导论》.北京:中国社会科学出版社. 1998.
[81]邬晓钧,郑方,徐明星.基于主题森林结构的对话管理模型.自动化学报, 29(2):275-283, 2003.
[82]吴立德.《大规模中文文本处理》.上海:复旦大学出版社. 1997.
[83]吴云芳.疑问句中的“怎么”在汉英机器翻译中的词汇歧义化解.辉煌二十年——中国中文信息学会二十周年学术会议论文集.北京. pp263-271, 2001.
[84]许嘉璐.现状和设想——试论中文信息处理与现代汉语研究.中国语文.第6期, pp1-8, 2000.
[85]徐烈炯.《语义学》.北京:语文出版社. 1990.
[86]颜国郎.口语对话系统中不流畅语音之语音动作型态模型化与验证之研究[博士学位论文].国立成功大学. 1993.
[87]袁保宗,吴佑寿.超级智能视听信息处理系统的研究.国家自然科学基金重大项目研究技术报告. 1993.12.
[88]詹人凤.《现代汉语语义学》.商务印书馆. 1997.
[89]朱学锋,俞士汶.自动翻译电话与口语信息处理研究.《人工智能新进展》,清华大学出版社, 1994.
[90]李家治,陈永明.机器理解自然语言中有关代词处理的几个问题.自然语言理解年会论文集. 1986.
[91]范继淹,徐志敏. RJD-80型汉语人机对话系统的语法分析.中国语文,第3期,pp223-232, 1982.
[92]黄寅飞,郑方,燕鹏举,徐明星,吴文虎.校园导航系统EasyNav的设计与实现.中文信息学报. 15(4):35-40, 2001.
[93]刘蓓,杜利民,于水源.面向任务口语对话系统中期待模型的实现算法.电子与信息学报.
26(11):1721-1727, 2004.
[94]张琳,高峰,郭荣,毛家菊,陆汝占.汉语股票实时行情查询对话系统.计算机应用. 24(7):61-63, 2004.
[95]郑杰,茅于杭,董清富.基于语境的语义排歧方法.中文信息学报. 14(5):1-8, 2000.
[96]周强,黄昌宁.汉语句法规则的自动构造方法研究.中文信息学报. 12(3):1-7, 1998.
[97]燕鹏举,陆正中,邬晓钧,徐明星等.航班信息系统EasyFlight.第六届全国人机语音通讯学术会议论文集. pp115-118, 2001.
[98]何伟,李红莲,袁保宗,林碧琴.基于对话回合衰减的cache语言模型在线自适应研究.中文信息学报. 17(5):41-47, 2003.
[99]阮秋琦,袁保宗.智能视听信息处理的新策略.北方交通大学学报, 20(2):152-160, 1996.