Abstract: Kowalski and Sergot's Event Calculus (EC) is a simple temporal formalism that, given a set of event occurrences, derives the maximal validity intervals (MVIs) over which properties initiated or terminated by these events hold. In this paper, we conduct a systematic analysis of EC by which we gain a better understanding of this formalism and determine ways of augmenting its expressive power. The keystone of this endeavor is the definition of an extendible formal specification of its functionalities. This formalization has the effects of casting determination of MVIs as a model checking problem, of setting the ground for studying and comparing the expressiveness and complexity of various extensions of EC, and of establishing a semantic reference against which to verify the soundness and completeness of implementations. We extend the range of queries accepted by EC, which is limited to Boolean combinations of MVI verification or computation requests, to support arbitrary quantification over events and modal queries. We also admit specifications based on preconditions. We demonstrate the added expressive power by encoding a number of diagnosis problems. Moreover, we provide a systematic comparison of the expressiveness and complexity of the various extended event calculi against each other. Finally, we propose a declarative encoding of these enriched event calculi in the logic programming language λProlog and prove the soundness and completeness of the resulting logic programs.

Abstract: In many areas of Computer Science, including planning, workflows, guidelines, and protocol management, one has to deal with abstract plans, procedures, or schedules involving temporal constraints between classes of actions that have to be repeated at periodic times and may be instantiated in different ways for different executions of the plans (procedures, schedules). In this paper we propose an integrated framework to deal with both qualitative temporal constraints on classes of actions and temporal constraints between instances of actions, in which temporal reasoning is used to amalgamate both types of constraints and to check their consistency. In particular, we consider an expressive formalism to deal with temporal constraints between classes of actions (with special attention to periodic actions) which takes into account different components such as frame times, numeric quantification, periods, and qualitative relations. We define the notions of (contextual) concretization of qualitative temporal constraints between classes and use this notion to formally define the consistency of a knowledge base of temporal constraints between classes and a set of temporal constraints on instances, and to define the algorithm for checking such a consistency. An application for scheduling lessons in a school is shown in an example.

Abstract: This work examines important issues in probabilistic temporal representation and reasoning using Bayesian networks (also known as belief networks). The representation proposed here utilizes temporal (or dynamic) probabilities to represent facts, events, and the effects of events. The architecture of a belief network may change with time to indicate a different causal context. Probability variations with time capture temporal properties such as persistence and causation. They also capture event interaction, and when the interaction between events follows known models such as the competing risks model, the additive model, or the dominating event model, the net effect of many interacting events on the temporal probabilities can be calculated efficiently. This representation of reasoning also exploits the notion of temporal degeneration of relevance due to information obsolescence to improve the efficiency.

Abstract: In the area of formal reasoning about action and change, one of the fundamental representation problems is providing concise modular and incremental specifications of action types and world models, where instantiations of action types are invoked by agents such as mobile robots. Provided the preconditions to the action are true, their invocation results in changes to the world model concomitant with the goal-directed behavior of the agent. One particularly difficult class of related problems, collectively called the qualification problem, deals with the need to find a concise incremental and modular means of characterizing the plethora of exceptional conditions that might qualify an action, but generally do not, without having to explicitly enumerate them in the preconditions to an action. We show how fluent dependency constraints together with the use of durational fluents can be used to deal with problems associated with action qualification using a temporal logic for action and change called TAL-Q. We demonstrate the approach using action scenarios that combine solutions to the frame, ramification, and qualification problems in the context of actions with duration, concurrent actions, nondeterministic actions, and the use of both Boolean and non-Boolean fluents. The circumscription policy used for the combined problems is reducible to the first-order case.

Abstract: This paper addresses the problem of granularity in temporal representation in the context of text analysis. In contrast with other fields where granularity levels are essentially quantitative, in natural language the different levels are not always precisely defined and granularity is of a more subtle and qualitative nature. A model is proposed for representing such phenomena, based on time units and time units intervals. Time units represent chunks of time which are considered indivisible at a given granularity level, but which may be refined and contain other time units or intervals at a higher granularity level. The structure of the set of time units and intervals is studied and a relation algebra is defined which extends the traditional Allen's Point or Interval Algebra. Weak and strong representations of this algebra are proposed. We claim that such representations, called grained temporal structures, are adequate for coping with dynamic qualitative changes of granularity. A logic with restricted quantifiers is proposed for formalizing temporal knowledge and examples are discussed which show the relevance of the model for natural language analysis.

Abstract: We propose a model, called a deliberation structure model, that represents streams of relevant messages in email communication. The deliberation structure model treats e-mail messages that include several topics and uses tree structure, called deliberation trees, that are based on Clark's contribution trees to represent message structure. Deliberation trees are useful to visualize the message structure. A method for constructing the deliberation structure is showed.

Abstract: This paper describes a system for the summarization of multiple text-only news-like documents We address two main issues: clustering of documents in order to find the main topics that should be mentioned in the multidocument summary and organization of the information in order to create a summary that presents the information in a logical way and is easy to read The system is based on a single-document summarizer that uses text-extraction

Abstract: We describe a prototype system, IKARUS, with which we investigated the potential of integrating web-based documents, shared knowledge bases, and information retrieval for improving knowledge storage and retrieval. As an example, we discuss how to implement both a user manual and an online help system as one system. The following technologies are combined: a web-based design, a frame-based knowledge engine, use of an advanced full-text search engine, and simple techniques to control terminology. We have combined graphical browsing with several unusual forms of text retrieval—for example, to the sentence and paragraph level.

Abstract: We discuss the use of probability-based natural language processing for Chinese text retrieval. We focus on comparing different text extraction methods and probabilistic weighting methods. Several docu- ment processing methods and probabilistic weighting functions are presented. A number of experiments have been conducted on large standard text collections. We present the experimental results that com- pare a word-based text processing method with a character-based method. The experimental results also compare a number of term-weighting functions including both single-unit weighting and compound-unit weighting functions. Modern information retrieval (IR) systems should be able to take natural language queries and retrieve documents written in natural languages. One way to process nat- ural language texts in IR is statistical. In this approach, linguistic units are extracted from the documents and queries. Those units extracted from the documents are used to index the documents. Those extracted from a query are weighted according to their degrees of importance. The most important units are chosen as query terms. An infor- mation retrieval system takes the selected query terms, matches the terms with the indexes of documents, calculates a retrieval status value for each matched document using a term-weighting method, and presents the user with potentially relevant docu- ments from a collection of documents. The performance of the information retrieval system in identifying relevant documents greatly depends on the text processing method used. We investigate the effectiveness of different text extraction methods and different term weighting methods in the context of Chinese information retrieval. It is a well- known problem that there is no separator between Chinese words, so Chinese words cannot be used easily to index or search texts as is possible in English. Therefore, some people use characters or n-grams as searchable tokens instead of words. We discuss two text extraction methods. One extraction method uses words, compound words, and phrases in the document and query texts as indexing terms to represent the texts. We refer to this method as the word-based approach. For this approach, text segmentation, which divides both document and query texts into linguistic units, is regarded not only as a necessary precursor but also as a bottleneck of this kind of system (Wu and Tseng 1993). The other extraction method is based on single Chinese characters, in which texts are indexed by the characters appearing in the texts (Chen 1992). By using sin- gle character approaches, a search could be conducted for any multi-character word or phrase identified at search time, no matter whether this word or phrase is in the dic- tionary. Both word-based and character-based methods have been used in information

Abstract: Time-series prediction is important in physical and financial domains. Pattern recognition techniques for time-series prediction are based on structural matching of the current state of the time-series with previously occurring states in historical data for making predictions. This paper describes a Pattern Modelling and Recognition System (PMRS) which is used for forecasting benchmark series and the US S&P financial index. The main aim of this paper is to evaluate the performance of such a system on noise free and Gaussian additive noise injected time-series. The results show that the addition of Gaussian noise leads to better forecasts. The results also show that the Gaussian noise standard deviation has an important effect on the PMRS performance. PMRS results are compared with the popular Exponential Smoothing method.

Abstract: The measurement of information is potentially as important to an information engineer as the measurement of physical quantities is to a civil or mechanical engineer. This article introduces semantic measures, representing the distance between the meanings of two messages. We demonstrate one possible application by giving a small-scale example in which a semantic measure was used to retrieve information by meaning rather than by word-occurrence. The distance function between the meanings of two messages can be generalized to cover fuzzy meanings. A possible application, the processing of free responses to opinion polls, is described.

Abstract: We propose Generate and Repair Machine Translation (GRMT), a constraint-based approach to machine translation (MT) that focuses on accurate translation output. The architecture of GRMT was designed to take advantage of, and have advantages over, the three classic strategies (Direct MT, Interlingual MT and Transfer MT), the nonlinguistic information strategies (Example-Based MT and Statistics-Based MT), and the hybrid strategies (Knowledge-Based MT and Shake-and-Bake MT) with respect to several translation aspects: simplicity, accuracy and multilingualism. GRMT performs the translation by generating a Translation Candidate (TC), verifying the syntax and semantics of the TC, and repairing the TC when required. GRMT comprises three modules: Analysis Lite Machine Translation (ALMT), Translation Candidate Evaluation (TCE), and Repair and Iterate (RI). In generating the TC, GRMT refines the scope of translation choices of each input word by taking into account the differences between languages in a unique way. In selecting an appropriate word for each input word, GRMT considers the semantic relationship between words. This semantic relationship is based on the Word Association (WordAsso) number. (WordAsso). WordAsso number is assigned to word class. Words are classified according to the meaning of words and their usage. Word classification is designed and used not only in the word selection process but also in the classifier selection process and in semantic representation. GRMT is highly modular and extendible in the following respects: each component is separated, not only the translation process components (ALMT, TCE, RI), but also in the knowledge-bases, each component can be extended easily to a larger domain. The adding of new languages is also possible since the source language (SL) and the target language (TL) are treated separately. The SL and the TL are connected via the SL-TL dictionary which contains simple information and is manageable. An English-Thai translation system has been implemented to illustrate the performance of GRMT. The system has been developed and run under SWI-Prolog 3.2.8. The English and Thai grammars have been developed based on the Head-Driven Phrase Structure Grammar (HPSG) and implemented on the Attribute Logic Engine (ALE). This English-Thai MT system was evaluated and it performs in the way we intended. ALMT generated acceptable translations (grammatically correct, correct word usage and convey the original meaning) for 47 out of the 90 sentences in the test corpus without repair. WE and RI improved 15 sentences using our current HPSG based grammars and lexicons. Twenty-one sentences which contain logical connections are first separated into linguistic units before the repair can be performed due to a current inadequacy in HPSG's semantic representation. However, each linguistic unit was then repaired successfully. Seven sentences faced with the problems of adding linking words and classifiers in Thai also require further research in order to develop ways to repair these sentences.

Abstract: Similarity in contextual behavior between words is considered a source of ‘lexical cohesion,’ which is otherwise hard to measure or quantify. Such contextual similarity is used by an implementation for text segmentation, the VecTile system, which uses precompiled vector representations of words to produce similarity curves over texts. The performance of this system is shown to improve over that of the TextTiling algorithm of Hearst (1997).

Abstract: In this paper, a new framework is proposed for defining the semantics of discourse relations and for defining the semantics of the utterances that relations link together. The proposal is to define relations in terms of the operation of an algorithm simulating the mental state of an agent interacting with the world. The algorithm interleaves perception, theorem proving, and action: The denotation of a complex utterance containing a relation between two simpler utterances is taken to be the description of the operation of the algorithm during the time interval identified by their Reichenbachian reference times. This proposal is presented in detail for two discourse relations. Its potential application in the treatment of mood, tense, aspect, and dialogue structure is also discussed in very general terms.

Abstract: Argument systems are based on the idea that one can construct arguments for propositions—structured reasons justifying the belief in a proposition. Using defeasible rules, arguments need not be valid in all circumstances, therefore, it might be possible to construct an argument for a proposition as well as its negation. When arguments support conflicting propositions, one of the arguments must be defeated, which raises the question of which (sub-) arguments can be subject to defeat. In legal argumentation, metarules determine the valid arguments by considering the last defeasible rule of each argument involved in a conflict. Since it is easier to evaluate arguments using their last rules, can a conflict be resolved by considering only the last defeasible rules of the arguments involved? We propose a new argument system where, instead of deriving a defeat relation between arguments, arguments for the defeat of defeasible rules are constructed. This system allows us to determine a set of valid (undefeated) arguments in linear time using an algorithm based on a JTMS, allows conflicts to be resolved using only the last rules of the arguments, allows us to establish a relation with Default Logic, and allows closure properties such as cumulativity to be proved. We propose an extension of the argument system based on a proposal for reasoning by cases in default logic.

Abstract: In this paper, we discuss how recent theoretical linguistic research focusing on the Minimalist Program (MP)(Cho95, Mar95, Zwa94)can be used to guide the parsing of a useful range of natural language sentences and the building of a logical representation in a principles-based manner. We discuss the components of the MP and give an example derivation. We then propose parsing algorithms that recreate the derivation structure starting with a lexicon and the surface form of a sentence. Given the approximated derivation structure, MP principles are applied to generate a logical form, which leads to linguistically based algorithms for determining possible meanings for sentences that are ambiguous due to quantifier scope. In this paper, we introduce a framework for describing the grammar of natural languages due to Noam Chomsky called the Minimalist Program (MP). We investigate how to build a parser that produces syntax trees conforming to the MP and how aspects of language that have bearing on meaning, but cannot be conveniently captured during parsing, can be processed. The linguistic framework is discussed first, followed by the computational implementation. We first give a brief overview of the Minimalist Program (Chomsky 1995; Marantz 1995; Zwart 1994). The MP is the latest incarnation of Principles and Parameters grammars that define language structure in terms of well-motivated principles and parameters that adapt the principles to various natural languages. However, in this paper, we do not describe a parser that implements the MP in all its cognitive implications, which are still not completely understood and are being investigated. We feel that it is a worthwhile exercise to use the basic principles of the MP to obtain rules and structures that enable the traditional implementation of a parser. Parsing, however, is only a part of the processing. There are many linguistic phenomena than can be handled only after a syntax tree has been obtained. We discuss, among other issues, further processing of the parse to handle issues in quantifier scoping. We think this part of the paper is interesting since it shows how vexing linguistic phenomena can be handled using simple computational techniques. In writing the parser, we use a set of sentence types that have been considered by those who have written parsers motivated by earlier versions of Principles and Parameters grammars. Merlo, in his paper on a parser based on an earlier version of Principles and Parameters grammar, called the Government and Binding Theory (GB), wrote that the set of sentences he chose constitutes a

Abstract: Using decision trees as a concept description language, we examine the time complexity for learning Boolean functions with polynomial-sized disjunctive normal form expressions when feature construction is performed on an initial decision tree containing only primitive attributes. A shortcoming of several feature-construction algorithms found in the literature is that it is difficult to develop time complexity results for them. We illustrate a way to determine a limit on the number of features to use for building more concise trees within a standard amount of time. We introduce a practical algorithm that forms a finite number of features using a decision tree in a polynomial amount of time. We show empirically that our procedure forms many features that subsequently appear in a tree and the new features aid in producing simpler trees when concepts are being learned from certain problem domains. Expert systems developers can use a method such as this to create a knowledge base of information that contains specific knowledge in the form of If-Then rules.

Abstract: A new framework for knowledge representation is proposed, based on modality-dependent (sight, sound, touch) images, within which agents will be better able to communicate using natural languages. The premise is that natural language processing depends not only on the grammatical rules but also on entities (whatever the agents are communicating about), their attributes, and supported events (in a time/space continuum). We call the collection of an entity's associated attributes a state. The description of a concept is refined when new information about the concept is encountered through discovery of consistency between the attributes belonging to different image instances.

Abstract: Effective manipulation of temporal information about periodic events is required for solving complex problems such as long-range scheduling or querying temporal information. Furthermore, many problems involving repeating events involve the optimization of temporal aspects of these events (e.g., minimizing make-span in job-shop scheduling). In this paper, a constraint-based formulation of reasoning problems with repeating events is presented, and its complexity is analyzed for a range of problems. Optimization constraints are interpreted formally using the Semiring CSPs (SCSP) representation of optimization in constraint reasoning. This allows for familiar algorithms such as branch-and-bound to be applied to solving them.

Abstract: A complete analysis of an English sentence includes syntactic, semantic, and pragmatic components. Presupposition belongs to the pragmatic component. How to determine the presuppositions of multiple-clause sentences has been the focus of much work. Projection of clausal presuppositions is one method to determine the presuppositions of multiple-clause sentences. In this paper we present a new approach to the projection problem. Drawing heavily on the theoretical techniques originating with Montague semantics, our system maps sentences of a category-based grammar into a set of expressions of intensional logic: one expression corresponding to the literal interpretation of the sentence and the remaining expressions corresponding to the presuppositions of the sentence. The new approach correctly predicts the presuppositions of a larger range of multiple-clause sentences than previous projection approaches.

Abstract: Recently, the prevalence of information retrieval engines has created an important application of the automatic summarization as the display of retrieval results, whereby the user can quickly and accurately judge the relevance of texts returned as a result of a query. Here, rather than producing a generic summary, the summary that reflects the user's topic of interest (information need) expressed in the query would be considered as more suitable. This type of summary is often called query-biased summary. In this paper we present a method for producing query-biased summaries using lexical chains. Lexical chains are sequences of words that are in lexical cohesion relation with each other, and tend to indicate fragments of a text that form a semantic unit. Using lexical chains would enable to produce more coherent and readable summaries than previous approaches to query-biased summarization. To evaluate the effectiveness of our method, a task-based evaluation scheme is adopted. The results from the experiments show that query-biased summaries by lexical chains outperform others in the accuracy of subjects' relevance judgments.

Abstract: This paper describes a machine learning approach to build an ecien t, accurate and fast name spotting system. Finding names in free text is an important task in addressing real-world textbased applications. Most previous approaches have been based on carefully hand-crafted modules encoding linguistic knowledge specic to the language and document genre. Such approaches have two drawbacks: they require large amounts of time and linguistic expertise to develop, and they are not easily portable to new languages and genres. This paper describes an extensible system which automatically combines weak evidence for name extraction. This evidence is gathered from easily available sources: part-of-speech tagging, dictionary lookups, and textual information such as capitalization and punctuation. Individually, each piece of evidence is insucien t for robust name detection. However, the combination of evidence, through standard machine learning techniques, yields a system that achieves performance equivalent to the best existing hand-crafted approaches.

Abstract: Although humans seem adept at drawing nonmonotonic conclusions, the nonmonotonic reasoning systems that researchers develop are complex and do not function with such ease. This paper explores people's reasoning processes in nonmonotonic problems. To avoid the problem of people's conclusions being based on knowledge rather than on some reasoning process, we developed a scenario about life on another planet. Problems were chosen to allow the systematic study of people's understanding of strict and nonstrict rules and their interactions. We found that people had great difficulty reasoning and we identified a number of negative factors influencing their reasoning. We also identified three positive factors which, if used consistently, would yield rational and coherent reasoning-but no subject achieved total consistency. (Another possible positive factor, specificity, was considered but we found no evidence for its use.) It is concluded that nonmonotonic reasoning is hard. When people need to reason in a domain where they have no preconceived ideas, the foundation for their reasoning is neither coherent nor rational. They do not use a nonmonotonic reasoning system that would work regardless of content. Thus, nonmonotonic reasoning systems that researchers develop are expected to do more reasoning than humans actually do!

Abstract: This paper discusses a fundamental problem in natural language generation: how to organize the content of a text in a coherent and natural way. In this research, we set out to determine the semantic content and the rhetorical structure of texts and to develop heuristics to perform this process automatically within a text generation framework. The study was performed on a specific language and textual genre: French instructional texts. From a corpus analysis of these texts, we determined nine senses typically communicated in instructional texts and seven rhetorical relations used to present these senses. From this analysis, we then developed a set of presentation heuristics that determine how the senses to be communicated should be organized rhetorically in order to create a coherent and natural text. The heuristics are based on five types of constraints: conceptual, semantic, rhetorical, pragmatic, and intentional constraints. To verify the heuristics, we developed the spin natural language generation system, which performs all steps of text generation but focuses on the determination of the content and the rhetorical structure of the text.

Abstract: The basic nearest-neighbor rule generalizes well in many domains but has several shortcomings, including inappropriate distance functions, large storage requirements, slow execution time, sensitivity to noise, and an inability to adjust its decision boundaries after storing the training data. This paper proposes methods for overcoming each of these weaknesses and combines the methods into a comprehensive learning system called the Integrated Decremental Instance-Based Learning Algorithm (IDIBL) that seeks to reduce storage, improve execution speed, and increase generalization accuracy, when compared to the basic nearest neighbor algorithm and other learning models. IDIBL tunes its own parameters using a new measure of fitness that combines confidence and cross-validation accuracy in order to avoid discretization problems with more traditional leave-one-out cross-validation. In our experiments IDIBL achieves higher generalization accuracy than other less comprehensive instance-based learning algorithms, while requiring less than one-fourth the storage of the nearest neighbor algorithm and improving execution speed by a corresponding factor. In experiments on twenty-one data sets, IDIBL also achieves higher generalization accuracy than that reported for sixteen major machine learning and neural network models.

Abstract: The possible models approach is a classical minimal change semantics for knowledge base update, which provides an exclusive interpretation for disjunctive information in updates. It has been recognized that the exclusive interpretation for disjunction may be problematic under some circumstances. In this paper, we investigate inclusive interpretations for disjunctions in updates from both syntactical and semantical viewpoints. In particular, we propose two approaches for disjunctive update—the minimal change with exceptions (MCE) and the minimal change with maximal disjunctive inclusions (MCD). Both approaches provide inclusive interpretations for disjunctions in updates, though the first is syntax-based and the second is model-theoretic. We then characterize the MCE and MCD in terms of alternative minimal change criteria and relate them to traditional Katsuno and Mendelzon's update postulates.