Showing papers on "Natural language understanding published in 2011"

Learning structured embeddings of knowledge bases

Abstract: Many Knowledge Bases (KBs) are now readily available and encompass colossal quantities of information thanks to either a long-term funding effort (e.g. WordNet, OpenCyc) or a collaborative process (e.g. Freebase, DBpedia). However, each of them is based on a different rigid symbolic framework which makes it hard to use their data in other systems. It is unfortunate because such rich structured knowledge might lead to a huge leap forward in many other areas of AI like natural language processing (word-sense disambiguation, natural language understanding, ...), vision (scene classification, image semantic annotation, ...) or collaborative filtering. In this paper, we present a learning process based on an innovative neural network architecture designed to embed any of these symbolic representations into a more flexible continuous vector space in which the original knowledge is kept and enhanced. These learnt embeddings would allow data from any KB to be easily used in recent machine learning methods for prediction and information retrieval. We illustrate our method on WordNet and Freebase and also present a way to adapt it to knowledge extraction from raw text.

Predicting user mental states in spoken dialogue systems

Abstract: In this paper we propose a method for predicting the user mental state for the development of more efficient and usable spoken dialogue systems. This prediction, carried out for each user turn in the dialogue, makes it possible to adapt the system dynamically to the user needs. The mental state is built on the basis of the emotional state of the user and their intention, and is recognized by means of a module conceived as an intermediate phase between natural language understanding and the dialogue management in the architecture of the systems. We have implemented the method in the UAH system, for which the evaluation results with both simulated and real users show that taking into account the user's mental state improves system performance as well as its perceived quality.

Building Timelines from Narrative Clinical Records: Initial Results Based-on Deep Natural Language Understanding

Abstract: We present an end-to-end system that processes narrative clinical records, constructs timelines for the medical histories of patients, and visualizes the results. This work is motivated by real clinical records and our general approach is based on deep semantic natural language understanding.

Integration of a voice recognition system in a social robot

Abstract: Human-robot interaction (HRI)1 is one of the main fields in the study and research of robotics. Within this field, dialogue systems and interaction by voice play an important role. When speaking about human-robot natural dialogue we assume that the robot has the capability to accurately recognize what the human wants to transmit verbally and even its semantic meaning, but this is not always achieved. In this article we describe the steps and requirements that we went through in order to endow the personal social robot Maggie, developed at the University Carlos III of Madrid, with the capability of understanding the natural language spoken by any human. We have analyzed the different possibilities offered by current software/hardware alternatives by testing them in real environments. We have obtained accurate data related to the speech recognition capabilities in different environments, using the most modern audio acquisition systems and analyzing not so typical parameters such as user age, gender, intonation, volume, and language. Finally, we propose a new model to classify recognition results as accepted or rejected, based on a second automatic speech recognition (ASR) opinion. This new approach takes into account the precalculated success rate in noise intervals for each recognition framework, decreasing the rate of false positives and false negatives.

Understanding the unknown: Unattested input processing in natural language

Abstract: This paper describes the process of deriving the meaning of an unknown word within the framework of meaning based natural language processing. It uses the clues supplied by the rest of the sentence, taking into account various degrees of possibilities of what the unknown word can mean, according to the previously acquired knowledge resources. The process of finding the meaning is incremental, and thus the derived meaning can evolve as more knowledge is gathered. The paper describes the fuzzy approach to the problem within this framework with regard to the semantic and syntactic acceptability of a sentence.

Opinion Mining and Sentiment Analysis Need Text Understanding.

Abstract: We argue in this paper that in order to properly capture opinion and sentiment expressed in texts or dialogs any system needs a deep linguistic processing approach. As in other systems, we used ontology matching and concept search, based on standard lexical resources, but a natural language understanding system is still required to spot fundamental and pervasive linguistic phenomena. We implemented these additions to VENSES system and the results of the evaluation are compared to those reported in the state-of-the-art systems in sentiment analysis and opinion mining. We also provide a critical review of the current benchmark datasets as we realized that very often sentiment and opinion is not properly modeled.

The Deep Lexical Semantics of Emotions

Abstract: The research described here is part of a larger effort, first, to construct formal theories of a broad range of aspects of commonsense psychology, including knowledge management, the envisionment of possible courses of events, and goal-directed behavior, and, second, to link them to the English lexicon. We have identified the most common words and phrases for describing emotions in English. In this paper we describe a formalization of people’s implicit theory of how emotions mediate between what they experience and what they do. We then sketch out effort to write rules that link the theory with words and phrases in the emotional lexicon.

Opinion Mining and Sentiment Analysis Need Text Understanding

Abstract: We argue in this paper that in order to properly capture opinion and sentiment expressed in texts or dialogs any system needs a deep linguistic processing approach. As in other systems, we used ontology matching and concept search, based on standard lexical resources, but a natural language understanding system is still required to spot fundamental and pervasive linguistic phenomena. We implemented these additions to VENSES system and the results of the evaluation are compared to those reported in the state-of-the-art systems in sentiment analysis and opinion mining. We also provide a critical review of the current benchmark datasets as we realized that very often sentiment and opinion is not properly modeled.

Towards the rapid development of a natural language understanding module

Abstract: When developing a conversational agent, there is often an urgent need to have a prototype available in order to test the application with real users. A Wizard of Oz is a possibility, but sometimes the agent should be simply deployed in the environment where it will be used. Here, the agent should be able to capture as many interactions as possible and to understand how people react to failure. In this paper, we focus on the rapid development of a natural language understanding module by non experts. Our approach follows the learning paradigm and sees the process of understanding natural language as a classification problem. We test our module with a conversational agent that answers questions in the art domain. Moreover, we show how our approach can be used by a natural language interface to a cinema database.

Natural Language Ontology of Action: A Gap with Huge Consequences for Natural Language Understanding and Machine Translation

Abstract: Action verbs are the less predictable linguistic type for bilingual dictionaries and they cause major problems for MT technologies that are immediately evident to the user This is not only because of language specific phraseology, but is rather a consequence of the peculiar way each natural language categorizes events ie it is a consequence of semantic factors In ordinary languages the most frequent Action verbs are “general”, since they extend productively to actions belonging to different ontological types Moreover, each language categorizes action in its own way and therefore the cross-linguistic reference to everyday activities is puzzling But the actual variation of verbs across action types is largely unknown This paper sketches the problem constituted by the Ontology of Action when disambiguation and cross-linguistic reference to action is concerned and presents the IMAGACT Ontology Infrastructure, which aims at filling this gap by exploiting multilingual spoken corpora

Efficient Graph Kernels for Textual Entailment Recognition

Abstract: One of the most important research area in Natural Language Processing concerns the modeling of semantics expressed in text. Since foundational work in Natural Language Understanding has shown that a deep semantic approach is still not feasible, current research is focused on shallow methods combining linguistic models and machine learning techniques. The latter aim at learning semantic models, like those that can detect the entailment between the meaning of two text fragments, by means of training examples described by specific features. These are rather difficult to design since there is no linguistic model that can effectively encode the lexico-syntactic level of a sentence and its corresponding semantic models. Thus, the adopted solution consists in exhaustively describing training examples by means of all possible combinations of sentence words and syntactic information. The latter, typically expressed as parse trees of text fragments, is often encoded in the learning process using graph algorithms. In this paper, we propose a class of graphs, the tripartite directed acyclic graphs (tDAGs), which can be efficiently used to design algorithms for graph kernels for semantic natural language tasks involving sentence pairs. These model the matching between two pairs of syntactic trees in terms of all possible graph fragments. Interestingly, since tDAGs encode the association between identical or similar words (i.e. variables), it can be used to represent and learn first-order rules, i.e. rules describable by first-order logic. We prove that our matching function is a valid kernel and we empirically show that, although its evaluation is still exponential in the worst case, it is extremely efficient and more accurate than the previously proposed kernels.

A general framework for incremental processing of multimodal inputs

Abstract: Humans employ different information channels (modalities) such as speech, pictures and gestures in their communication. It is believed that some of these modalities are more error-prone to some specific type of data and therefore multimodality can help to reduce ambiguities in the interaction. There have been numerous efforts in implementing multimodal interfaces for computers and robots. Yet, there is no general standard framework for developing them. In this paper we propose a general framework for implementing multimodal interfaces. It is designed to perform natural language understanding, multi- modal integration and semantic analysis with an incremental pipeline and includes a multimodal grammar language, which is used for multimodal presentation and semantic meaning generation.

A model based framework for semantic interpretation of architectural construction drawings

Abstract: The study addresses the automated translation of architectural drawings from 2D Computer Aided Drafting (CAD) data into a Building Information Model (BIM), with emphasis on the nature, possible role, and limitations of a drafting language Knowledge Representation (KR) on the problem and process. The central idea is that CAD to BIM translation is a complex diagrammatic interpretation problem requiring a domain (drafting language) KR to render it tractable and that such a KR can take the form of an information model. Formal notions of drawing-as-language have been advanced and studied quite extensively for close to 25 years. The analogy implicitly encourages comparison between problem structures in both domains, revealing important similarities and offering guidance from the more mature field of Natural Language Understanding (NLU). The primary insight we derive from NLU involves the central role that a formal language description plays in guiding the process of interpretation (inferential reasoning), and the notable absence of a comparable specification for architectural drafting. We adopt a modified version of Engelhard’s approach which expresses drawing structure in terms of a symbol set, a set of relationships, and a set of compositional frameworks in which they are composed. We further define an approach for establishing the features of this KR, drawing upon related work on conceptual frameworks for diagrammatic reasoning systems. We augment this with observation of human subjects performing a number of drafting interpretation exercises and derive some understanding of its inferential nature therefrom. We consider this indicative of the potential range of processes a computational drafting model should ideally support. The KR is implemented as an information model using the EXPRESS language because it is in the public domain and is the implementation language of the target Industry Foundation Classes (IFC) model. We draw extensively from the IFC library to demonstrate that it can be applied in this manner, and apply the MVD methodology in defining the scope and interface of the DOM and IFC. This simplifies the IFC translation process significantly and minimizes the need for mapping. We conclude on the basis of selective implementations that a model reflecting the principles and features we define can indeed provide needed and otherwise unavailable support in drafting interpretation and other problems involving reasoning with this class of diagrammatic representations.

Intrinsic and Extrinsic Approaches to Recognizing Textual Entailment

Abstract: Recognizing Textual Entailment (RTE) is to detect an important relation between two texts, namely whether one text can be inferred from the other. For natural language processing, especially for natural language understanding, this is a useful and challenging task. We start with an introduction of the notion of textual entailment, and then define the scope of the recognition task. We summarize previous work and point out two important issues involved, meaning representation and relation recognition. For the former, a general representation based on dependency relations between words or tokens is used to approximate the meaning of the text. For the latter, two categories of approaches, intrinsic and extrinsic ones, are proposed. The two parts of the thesis are dedicated to these two classes of approaches. Intrinsically, we develop specialized modules to deal with different types of entailment; and extrinsically, we explore the connection between RTE and other semantic relations between texts. In the first part, an extensible architecture is presented to incorporate different specialized modules handling different types of entailment. We start with one specialized module for handling text pairs with temporal expressions. A separate time anchoring component is developed to recognize and normalize the temporal expressions contained in the texts. Then it is shown that the generalization of this module can handle texts containing other types of named-entities as well. The evaluation results confirm that precision-oriented specialized modules are required. We also describe another module based on an external knowledge resource. A collection of textual inference rules is applied to the RTE task after being extended and refined with a hand-crafted lexical resource. The evaluation results demonstrate that this is a precision-oriented approach, which can also be viewed as a specialized module. As alternative resources, we also present a pilot study on acquiring paraphrased fragment pairs in an unsupervised manner. In the second part of the dissertation, a general framework is proposed to view textual entailment as one of the generalized Textual Semantic Relations (TSRs). Instead of tackling the RTE task in a standalone manner, we look at its connection to other semantic relations between two texts, e.g., paraphrase, contradiction, etc. The motivation of such a generalization is given as well as the framework of recognizing all these

Accessing Structured Health Information through English Queries and Automatic Deduction

Abstract: While much health data is available online, patients who are not technically astute may be unable to access it because they may not know the relevant resources, they may be reluctant to confront an unfamiliar interface, and they may not know how to compose an answer from information provided by multiple heterogeneous resources. We describe ongoing research in using natural English text queries and automated deduction to obtain answers based on multiple structured data sources in a specific subject domain. Each English query is transformed using natural language technology into an unambiguous logical form; this is submitted to a theorem prover that operates over an axiomatic theory of the subject domain. Symbols in the theory are linked to relations in external databases known to the system. An answer is obtained from the proof, along with an English language explanation of how the answer was obtained. Answers need not be present explicitly in any of the databases, but rather may be deduced or computed from the information they provide. Although English is highly ambiguous, the natural language technology is informed by subject domain knowledge, so that readings of the query that are syntactically plausible but semantically impossible are discarded. When a question is still ambiguous, the system can interrogate the patient to determine what meaning was intended. Additional queries can clarify earlier ones or ask questions referring to previously computed answers. We describe a prototype system, Quadri, which answers questions about HIV treatment using the Stanford HIV Drug Resistance Database and other resources. Natural language processing is provided by PARC’s Bridge, and the deductive mechanism is SRI’s SNARK theorem prover. We discuss some of the problems that must be faced to make this approach work, and some of our solutions.

Actions Speak Louder than Words: Evaluating Parsers in the Context of Natural Language Understanding Systems for Human-Robot Interaction

Abstract: The standard ParsEval metrics alone are often not sufficient for evaluating parsers integrated in natural language understanding systems. We propose to augment intrinsic parser evaluations by extrinsic measures in the context of human-robot interaction using a corpus from a human cooperative search task. We compare a constituent with a dependency parser on both intrinsic and extrinsic measures and show that the conversion to semantics is feasible for different syntactic paradigms.

Lattice-based approach to building templates for natural language understanding in intelligent tutoring systems

Abstract: We describe a domain-independent authoring tool, ConceptGrid, that helps non-programmers develop intelligent tutoring systems (ITSs) that perform natural language processing. The approach involves the use of a lattice-style table-driven interface to build templates that describe a set of required concepts that are meant to be a part of a student's response to a question, and a set of incorrect concepts that reflect incorrect understanding by the student. The tool also helps provide customized just-in-time feedback based on the concepts present or absent in the student's response. This tool has been integrated and tested with a browser-based ITS authoring tool called xPST.

Object-oriented semantics of English in natural language understanding system

Abstract: A new approach to the problem of natural language understanding is proposed The knowledge domain under consideration is the social behavior of people English sentences are translated into set of predicates of a semantic database, which describe persons, occupations, organizations, projects, actions, events, messages, machines, things, animals, location and time of actions, relations between objects, thoughts, cause-and-effect relations, abstract objects There is a knowledge base containing the description of semantics of objects (functions and structure), actions (motives and causes), and operations

Understanding in distributed social-administrative network systems

Abstract: Understanding is crucial for smart man-machine communication and therefore in the future of a well balanced information society. A pragmatic definition is given, valid for machines and humans. The solutions follow the developmental course in the history of philosophy and realize them based on the pattern concept of information science and the intelligent methods of frames, scripts, semantic nets and natural language understanding achievements. An extensive experiment using corpora of citizens claims indicate the feasibility and limits in application of current information technology and the emphasis of the elevated human role.

Question Answering in a Natural Language Understanding System Based on Object-Oriented Semantics

Abstract: Algorithms of question answering in a computer system oriented on input and logical processing of text information are presented. A knowledge domain under consideration is social behavior of a person. A database of the system includes an internal representation of natural language sentences and supplemental information. The answer {\it Yes} or {\it No} is formed for a general question. A special question containing an interrogative word or group of interrogative words permits to find a subject, object, place, time, cause, purpose and way of action or event. Answer generation is based on identification algorithms of persons, organizations, machines, things, places, and times. Proposed algorithms of question answering can be realized in information systems closely connected with text processing (criminology, operation of business, medicine, document systems).

An Evaluation of Alternative Strategies for Implementing Dialogue Policies Using Statistical Classification and Rules

Abstract: We present and evaluate a set of architectures for conversational dialogue systems, exploring rule-based and statistical classification approaches. In a case study, we show that while a rule-based dialogue policy is capable of high performance if perfect natural language understanding is assumed, a direct classification approach that combines the dialogue policy with NLU has practical advantages.

Why we need evolutionary semantics

Abstract: One of the key components for achieving flexible, robust, adaptive and open-ended language-based communication between humans and robots - or between robots and robots - is rich deep semantics. AI has a long tradition of work in the representation of knowledge, most of it within the logical tradition. This tradition assumes that an autonomous agent is able to derive formal descriptions of the world which can then be the basis of logical inference and natural language understanding or production. This paper outlines some difficulties with this logical stance and reports alternative research on the development of an 'embodied cognitive semantics' that is grounded in the world through a robot's sensori-motor system and is evolutionary in the sense that the conceptual frameworks underlying language are assumed to be adapted by agents in the course of dialogs and thus undergo constant change.

Intelligent Tutoring of School Level Geometry Using Automatic Text to Diagram Conversion Utility

Abstract: This paper reports a novel text-to-diagram conversion mechanism that simulates human approach of solving a geometry problem stated in English. Thereby it establishes an implement-able framework towards developing intelligent CBT tool for school-level geometry problems. Realization of the framework involves building of a geometry knowledge base and application of many aspects of natural language understanding.

Symbolic Probabilistic Reasoning for Narratives

Abstract: We present a framework to represent and reason about narratives that combines logical and probabilistic representations of commonsense knowledge. Unlike most natural language understanding systems, which merely extract facts or semantic roles, our system builds probabilistic representations of the temporal sequence of world states and actions implied by a narrative. We use probabilistic actions to represent ambiguities and uncertainties in the narrative. We present algorithms that take a representation of a narrative, derive all possible interpretations of the narrative, and answer probabilistic queries by marginalizing over all the interpretations. With a focus on spatial contexts, we demonstrate our framework on an example narrative. To this end, we apply natural language pro- cessing (NLP) tools together with statistical approaches over common sense knowledge bases.

Designing service-oriented chatbot systems using a construction grammar-driven natural language generation system

Abstract: Service oriented chatbot systems are used to inform users in a conversational manner about a particular service or product on a website. Our research shows that current systems are time consuming to build and not very accurate or satisfying to users. We find that natural language understanding and natural language generation methods are central to creating an e�fficient and useful system. In this thesis we investigate current and past methods in this research area and place particular emphasis on Construction Grammar and its computational implementation. Our research shows that users have strong emotive reactions to how these systems behave, so we also investigate the human computer interaction component. We present three systems (KIA, John and KIA2), and carry out extensive user tests on all of them, as well as comparative tests. KIA is built using existing methods, John is built with the user in mind and KIA2 is built using the construction grammar method. We found that the construction grammar approach performs well in service oriented chatbots systems, and that users preferred it over other systems.

Natural Language Understanding, Where Are We Going? Where Could We Go?

Abstract: Historically, at the beginning of natural language processing, applications with industrial objectives were preferred, e.g. for a fully automated translation. Soon, during the 1960s, there were attempts to separate applications from basic theoretical research. Since this theoretical research encounters difficulties regarding its status as a separate discipline, and considering the constraints necessary to clarify the concept of ‘good application’, impossible to satisfy simultaneously (a real problem to solve, in response to a social demand, and a viable solution in terms of reliability, robustness, speed and cost), two main streams have emerged. The first one, as a computer technique, is intended to build applications based on a strict logic, using natural language to facilitate interaction with the computer, but not directly related to the human way of using language (this approach is designated as natural language processing). Such pragmatic research accepts certain kinds of errors, but must lead to concrete results in limited time. The goal is to provide effective systems for real applications, able to respond effectively to requests addressed to them in fairly large areas; these systems are directly related to social and industrial productivity, which is the essential criterion of evaluation. Some technological developments, such as microcomputers, have made available to people specific applications of natural language processing and have enabled the emergence of small specialized firms. This produced, in the second half of the 1980s, the emergence of a ‘language industry’ and of the field of ‘linguistic engineering’. On the other hand, during the late 1960s, the gap between the social demand, the resources invested and the poor performance obtained led to the emergence of theoretical studies intended to formalize languages (as opposed to the more empirical machine translation). This leads to ‘pilot systems’, aimed at demonstrating the feasibility of complex theoretical approaches, but unable to operate outside a set of rather limited examples. The limits may be at different levels: more or less limited vocabulary or accepted sentences, knowledge about the field more or less complete, more or less developed reasoning and so on. These limits have a significant impact on communication itself. For natural language processing systems to be effective, they must make appropriate inferences from what is said and, conversely their behavior should allow the inferences that the users usually do when using their language. Thus, this position paper stresses that understanding the surface meaning of a natural language is not sufficient but that the goals, intentions and strategies of the participants in a dialogue must be understood.

A semantic processing model for sentence understanding based on cognitive learning

Abstract: Sentence has a very prominent position in the research field of Natural Language Understanding. The task of sentence understanding includes two stages, sentence parsing and semantic processing. Sentence parsing resides in the fundamental level, while semantic understanding involves lexcial and higher discourse analysis. As sentence understanding has compact connections with human cognition, this paper will introduce how cognitive models are integrated, with machine learning algorithms (or models), into the procedures of sentence parsing and semantic processing.