Artificial Intelligence Techniques

Download Full-Text PDF Cite this Publication

Text Only Version

Artificial Intelligence Techniques

Artificial Intelligence Techniques

Rishita Jain, Divya Nahata, Priyanka Mangal Department of Computer Science and Technology Mandsaur Institute Of Technology, Mandsaur, India

rishitajain118@gmail.com, Divyanahata786@gmail.com, priyanka.mangal@mitmandsaur..info

Abstract:-Artificial Intelligence (AI) involves two basic ideas. First, it involves studying the thought processes of human beings. Second, it deals with representing those processes via machines (computers, robots, etc). AI is the behavior of a machine, which, if performed by a human being, would be called intelligent. It makes machines smarter and more useful, is less expensive than natural intelligence. Natural Language Processing (NLP) refers to Artificial Intelligence methods of communicating with a computer in a natural language like English. The main objective of a NLP program is to understand input and initiate action. The input words are scanned and matched against internally stored known words. Identification of a keyword causes some action to be taken. In this way, one can communicate with computer in one language. AI methods are applied for speech recognition. Speech recognition (in many contexts, also known as automatic speech recognition, computer speech recognition or voice recognition) is the process of converting a speech signal to a set of words, by means of an algorithm implemented as a computer program.

Keywords:-Artificial Intelligence, space region

  1. INTRODUCTION:-

    AI is often defined in terms of the problems it studies. But in fact, AI is not the study ofintelligent behavior etc., it is a way to study it. This is evident from the way AI is done inpractice. This paper illustrates this by contrasting AI and alternative approaches to speech understanding. By so doing it brings out some key characteristics of AI methodology.This paper is primarily written for a general AI audience interested in methodologicalissues, complementing previous work (Cohen 1991; Brooks 1991a). It is also written for any AI researchers who are contemplating starting a project in speech understanding it is intended to be the paper that, if available earlier, might have saved me four years of wasted e_ort. This paper does not provide technical discussion of specific issues in speech understanding; for that, see (Ward 1996) and the references given below. The points made here are based on, first, a review of the full spectrum of published work on speech understanding, including AI work from the 1970s to the present (Reddy et al) Acknowledgements: Thanks to Dan Jurafsky, Jane Edwards, and Wataru Tsukahara for comments,

    And to the Sound Technology Promotion Foundation for support. 1973; Woods & Makhoul 1973; Klatt 1977; Erman et al. 1980; Woods 1980; Hayes et al.1987; Kitano et al. 1989; Baggia & Rullent 1993; Nagao et al. 1993; Kawahara et al. 1994;Hauenstein & Weber 1994; Cochard & Oppizzi 1995; Weber & Wermter 1996), and also engineering work, including (Moore et al. 1989; Lee 1994; Nguyen et al. 1994; Moore 1994; Alshawi & Carter 1994; Hirschman 1994; Levin & Pieraccini 1995; Jurafsky et al. 1995;Sene_ 1995; Moore et al. 1995; Pallett & Fiscus 1995) and psycholinguistic work, including (McClelland 1987; Norris 1993; Nygaard & Pisoni 1995; Cutler 1995); and second, my own experience building an AI speech understanding system (Ward 1992;Ward 1993;Ward 1994a; Ward 1994b; Ward 1995). The discussion is general, abstract, and simplistic: no specific project is characterized accurately, there is no attempt to distinguish among the various AI approaches, and there is no discussion of approaches part way along the continuum between AI and rival approaches. The aim is to bring together some diverse observations; no individual point is claimed to be original.

  2. CHARACTERISTICS OF AI APPROACH

This section discusses the AI approach in terms of four key values of AI: ambitious goals,introspective plausibility, computational elegance, and wide significance. These remarks apply specifically to classical AI in its pure form, but are also more generally relevant.

  1. Ambitious Goals and Bold Leaps

    AI research, like AI research more generally, motivated not by what can be achieved soon, but by a long-term vision. For speech, this has been the idea of a system able to produce an optimal interpretation based on exhaustive processing of the input.Engineers prefer to set goals towards which progress can be measured. AI speech research tends, since existing systems are nowhere near optimal, to seek breakthroughs and radically new perspectives.Engineers tend to proceed by improving existing systems. AI speech research, like other AI research, tries to solve problems in their most general form, often by trying to construct a single, general-purpose

    system.Engineering is in large part the creation of solutions for specific problems.consulting accumulation of know-how useful for solving other problems.

  2. INTROSPECTIVE PRAUSBILITY

    AI speech research, like AI more generally, often makes recourse to introspection about human abilities. This subsection illustrates the use of introspection at four phases: when setting long-term goals, when setting short-term goals, for design, and when debugging.

    1. When setting long-term research goals, AI speech research, like AI research in general, aims to duplicate apparent human abilities. In particular, many AI speech researchers are inspired by the apparent near perfection of human speech understanding: the fact that people can understand just about anything you say to them, and even repeat back to you the words you said.Scientific approaches to human speech understanding, in contrast, find it more productiveto focus on what can be learned from the limitations of human speech understanding. (A simple demonstration of these limitations can be had by simply recording a conversation and later listening to it repeatedly; you will discover that you missed a lot when hearing it live.) In general, the feeling that perception and understanding is complete is an illusion of introspection (Brooks 1991a; Dennett 1991).

    2. When setting short-term goals, AI speech research, like AI more generally, values systems which do things which seem, introspectively, clever. Such cleverness is often found not in the overall performance but in the details of processing in specific cases. For speech understanding, AI has emphasized the use of reasoning to compensate for failures of the lowlevel recognition component, often by selecting or creating word hypotheses for which therecognizer had little or no bottom-up evidence. Doing this can involve arbitrarily obscure knowledge and arbitrarily clever inferences, which makes for impressive traces of system operation. Engineers typically design and tune systems to work well on average; they seldom showo_ cleverness in specific cases. Few engineered speech systems do much explicit reasoning, and none bother to explicitly correct mis-recognitions rather, they simply barge on to compute the most likely semantic interpretations.

    3. For design, AI speech research, like AI more generally, takes inspiration from human cognitive architecture, as revealed by introspection. For

      speech, this has led to the use of protocol studies, done with a committee or with one man and a scratchpad. Both suggest a view of speech understanding as problem solving, and suggest a process where diverse knowledge sources cooperate by taking turns for example, with a partially recognized input leading to a partial uderstanding, that understanding being used to figure outmore words, leading to a better recognition result, then a better understanding, and so on.This has been fleshed out into, for example, blackboard models, which include a focus on explicitses, a focus on the interaction of knowledge sources, afocus on the scheduling of processing, and an image of iterative refinement over many cycles of computation.Scientific approaches, on the other hand, do not consider introspection to be reliable or complete. For example, introspection can easily focus on conscious figuring out, explicit decisions, and the role of grammar, but probably not on automatic processing, parallelprocessing of heses, and the role of prosody.

    4. . For development, AI speech research, like AI more generally, values systems which people can understand intuitively. makes it possible to debug by examining behavior on specific cases and adjusting the system until it works in a way that introspectively seems right. By doing so the developer can see if the system is a proper implementation of his vision. More important, he can get a feel for whether that vision is workable. In other words, the developer can use experience with the internal workings of a program to leverage introspection about how a cognitive process might operate. This technique is perhaps the most distinctive aspect of AI methodology; it could be called the hackers path to intelligent systems or perhaps computational introspection. Engineers focus on the desired input-output behavior of a system and design algorithms toachieve it. They typically do not care about the introspective plausibility of the intermediate steps of the computation.

  3. COMPUTATIONAL ELEGANCE

    AI speech research, like AI research more generally, postulates that knowledge is good and that more knowledge is better. In order to bring more knowledge to bear on specific decisions, integration of knowledge sources is considered essential. For speech, this means, most typically, wanting to use the full inventory of higher-level knowledge, including

    knowledge of syntax, semantics, domain, task and current dialog state, at the earliest stages of recognition and understanding. Engineers focus on performance, for which more knowledge may or may not be worthwhile, especially when processing time and memory are finite. Whether a specific type of

    knowledge is worthwhile, and if so when to apply it, are determined by experiment. AI speech research, like other AI research, involves an aesthetic sense of what designs are good. In particular, to build systems that can be extended and scaled up, AI researchers

    generally feel that elegant designs are required. For speech, the meaning of elegant has changed along with broader fashions in computer science at various times it has included: explicit (declarative) representation of hypotheses and control structures, emergent properties, multiprocessor (distributed) parallelism, fine-grained (connectionist, massive) parallelism, uniformity, heterogeneity, symbolic reasoning, evidential reasoning, and so on.Engineers prefer to actually try to build big systems, rather than just build prototypes and argue that they will scale up.

  4. WIDE SIGNIFICANCE

AI speech research, like AI research in general, has a larger purpose: researchers dont just want to solve the problem at hand, they also want their solution to inspire other work. AI speech researchers have generally wanted their work to be relevant to other problems in natural language processing and to linguistics, and have sought out and focused on phenomena of interest to those fields, such as ambiguity.AI speech researchers have also wanted to be relevant for the larger AI community. They have emphasized analogies relating speech understanding to other topics, such as search and planning. They have also emphasized ties to general AI issues, such as the Attention Problem,the Constraint Propagation Problem, the Context Problem, the Disambiguation Problem, the Evidential Reasoning Problem, the Knowledge Integration Problem, the Knowledge Representation Problem, the Noisy Input Problem, the Real-World Problem, the Reasoning with Uncertainty Problem, the Sensor Fusion Problem, the Signal- to-Symbol Problem, and a few others.AI speech researchers have also tried to make their work relevant for computer science more generally. Based on insights from speech understanding they have called for new architectures for computer networks, for software systems, and for computer hardware.Engineers prefer to work on goals rather than

interesting problems. They also preferto work on speech for its own sake, rather than for what it reveals about other problems.They tend to think in terms of real problems, not abstract ones.

    1. OUTCOME

      For speech understanding, and more generally, AI approaches have seemed more promising than traditional science and engineering. This is probably because AI methodology exploits introspection (§2.2) and aesthetic considerations (§2.3), both of which seem to provide the right answers with relatively little e_ort.However AI methodology has not fulfilled this promise for speech. The engineering approach, in contrast, has produced solid and impressive results.The significance of this is not obvious. Some AI researchers believe that this means that the speech community should be

      welcomed back to AI, as argued by several recent editorials. But the values of AI and mainstream (engineering- style) speech research are sodi_erent, as seen in §2, that reconciliation does not seem likely.Other AI researchers take the view that the results of engineering work are not interesting; presumably meaning that they are compelling neither introspectively or aesthetically. Many further believe that the AI approach to speech will be vindicated in the end. A few strive towards this goal (I was one). However, AI goals conflict with other more important goals, and so it is hard to be optimistic about future attempts to apply AI methodology to the speech understanding problem.The conclusions of this case study are thus in line with the conclusions of Brooks case study (Brooks 1991a). For both speech understanding and robotics, AI methodology turns out to be of little value. Whether this is also true in other cases is a question of great interest.

    2. LARGER SIGNIFICANCE

      The values of AI outlined in §2 actually best characterize the methodology of classical AI,ascendent in the 1970s, but less popular in recent years, with the proliferation of variant approaches to AI, including some, such as connectionism and Brooks-style1 AI (Brooks 1991a;Brooks 1991b), which reject many of the values of mainstream AI. Nevertheless, classical AI methodology still has wide influence.It is interesting to note that adopting a Brooks-style approach to speech leads to the view that, for building spoken dialog systems, it is not speech understanding that is the key. For one thing, the values of §2 were important when the topics and tasks of AI were beingestablished. As a result, those working on knowledge representation, planning, reasoning, distributed intelligence, user modeling, etc. today, even if they do not share the classical values of AI, are working on problems posed by researchers who did. Moreover, despite changes in AI methodology (mostly in the direction of methodological respectability, in the sense of importing and techniques from engineering and science), the values of §2 are not only still alive, but remain intrinsic to the existence of AI as a field. It is true that, at recent AI conferences, most papers do not

      explicitly refer to these values, but the fact that the papers appear at AI conferences at all is tribute to them. If AI researchers did not rely on introspection, have grand goals, or value aesthetic considerations, most of them would drift o_ to conferences on computational psychology, user interface engineering, specific applications, etc. And without the belief that results sould have wide significance and utility, groups of AI researchers would drift o_ to separate fields of speech, vision, motion,etc. In any case, it is clear that AI is a distinctive field of study in many ways. This paper has been an endeavor to pinpoint just what it is that is special about AI.

    3. CONCLUSION

      AI is often defined in terms of the problems it studies. But in fact, AI is not the study ofintelligent behavior etc., it is a way to study it. This is evident from the way AI is done inpractice. This paper illustrates this by contrasting AI and alternative approaches to speech understanding. By so doing it brings out some key characteristics of AI methodology.This paper is primarily written for a general AI audience interested in methodologicalissues, complementing previous work (Cohen 1991; Brooks 1991a). It is also written for any AI researchers who are contemplating starting a project in speech understanding it is intended to be the paper that, if available earlier, might have saved me four years of wasted e_ort. This paper does not provide technical discussion of specific issues in speech understanding; for that, see (Ward 1996) and the references given below. The points made here are based on, first, a review of the full spectrum .

  1. Bates, The Nature of Character in Interactive Worlds and The Oz Project, Technical Report CMU-CS-92-200, School Of Computer Science, Carnegie Mellon University, Pittsburgh, PA. October 1992.

  2. Forbus, K. Nielsen, P. and Faltings, B., Quailed Spatial Reasoning: The CLOCK Project, Artificial Intelligence, 51(1 3), October 1991.

  3. Coco, D., Creating Intelligent Creatures, Computer Graphics World, July 1997.

  4. Forbus, K. D., Mahoney, J. V. and Dill, K., How Qualitive Spatial Reasoning Can Improve Strategy Game AIs, Proceedings of the 2001 AAAI Spring Symposium on Artificial Intelligence and Computer Games, 2000.

  5. Gibson, C. and OBrien, J., the Basics of Team AI, Proceedings of the 2001 Game Developers Conference, pp. 323 332, 2001.

  6. Kerkez, B., Cox, M. T., Incremental Case-Based Plan Recognition Using State Indices, August, 2001.

  7. Laird, J. E., van Lent, M., Human-level AIs Killer Application: Interactive Computer Games, Proc. AAAI 2000, pp. 1171-1178. AAAI Press / the MIT Press, 2000.

  8. Aamodt, A., and Plaza, E. 1994, Case-based reasoning. Foundational issues, methodological variations, and system Approaches, Artificial Intelligence Communications 7(1): 3959.

Leave a Reply

Your email address will not be published. Required fields are marked *