Semantic Text Analysis Research Articles

Producing rules for ascertaining meanings of minimal semantic-syntactic text units

The task of ascertaining minimal semantic-syntactic text units within the framework of the task of the semantic analysis of texts in the Russian language is considered. The use of the DSM-method for producing rules for ascertaining the meanings of syntaxemes is described, the setting of the goal of producing rules in DSM-method terminology is presented. The analysis of the results of computer experiments and the causes of the presence of contradictory rules are given.

Toward Automatic Determinationof the Semantics of Connectivesin Large Newspaper Corpora

We explored the possibility of using automatic techniques to analyze the use of backward causal connectives in large Dutch newspaper corpora. With the help of 2 techniques, Latent Semantic Analysis and Thematic Text Analysis, the contexts of more than 14,000 connectives were studied. The method of analysis is described. We found that differences that have been suggested in the literature via hand-based analyses between these types of connectives (e.g., on dimensions such as subjectivity, change in perspective, and factuality of the connected segments) also appear in our corpus of 16.5 million words.

Editorial—A Survey of Research Questions for Intelligent Information Systems in Education

Education is an application domain in which many research questions from Intelligent Information Systems may prove their worth. We discuss three themes in this editorial: distributed education and learner modeling, semantic analysis of text, and intelligent information management.

Profiling large-scale lazy functional programs

The LOLITA natural language processor is an example of one of the ever-increasing number of large-scale systems written entirely in a functional programming language. The system consists of over 47,000 lines of Haskell code (excluding comments) and is able to perform a wide range of tasks such as semantic and pragmatic analysis of text, information extraction and query analysis. The efficiency of such a system is critical; interactive tasks (such as query analysis) must ensure that the user is not inconvenienced by long pauses, and batch mode tasks (such as information extraction) must ensure that an adequate throughput can be achieved. For the past three years the profiling tools supplied with GHC and HBC have been used to analyse and reason about the complexity of the LOLITA system. There have been good results, however experience has shown that in a large system the profiling life-cycle is often too long to make detailed analysis possible, and the results are often misleading. In response to these problems a profiler has been developed which allows the complete set of program costs to be recorded in so-called cost-centre stacks. These program costs are then analysed using a post-processing tool to allow the developer to explore the costs of the program in ways that are either not possible with existing tools or would require repeated compilations and executions of the program. The modifications to the Glasgow Haskell compiler based on detailed cost semantics and an efficient implementation scheme are discussed. The results of using this new profiling tool in the analysis of a number of Haskell programs are also presented. The overheads of the scheme are discussed and the benefits of this new system are considered. An outline is also given of how this approach can be modified to assist with the tracing and debugging of programs.

3. A Generic Semantic Grammar for Quantitative Text Analysis: Applications to East and West Berlin Radio News Content from 1979

In a semantic text analysis the researcher begins by creating one of two types of semantic grammars, each of which provides one or more templates that specify the ways concepts (or more general themes) may be related. On the one hand, a phenomenal semantic grammar can be created to extract phenomenon-related information from a text population (e.g., “Among the population's grievances [the phenomenon of interest in this case], which were ones for the abolition of taxes?”). On the other hand, a generic semantic grammar may be developed to yield data about the text population itself (e.g., “Among all clauses in the text population, how many were grievances for the abolition of taxes?”). This paper describes a generic semantic grammar that can be used to encode themes and theme relations in every clause within randomly sampled texts. Unlike the surface-grammatical relations mapped by syntax grammars, the theme relations allowed in this grammar only permit unambiguous encoding according to the meanings that clauses were intended to convey within their social context. An application of the grammar provides a concrete illustration of its research potential

Training the Translator

This book begins by investigating, through the use of think-aloud protocols, the mental processes of students when they translate. The creative and successful processes observed can be used directly for teaching purposes, while the unsuccessful ones can serve to find out where remedial training is needed. The book then goes on to discuss methods for improving a translator's competence. The strategies offered are based on the pragmatic and semantic analysis of texts from a functional point of view, and they include such practical matters as the use of dictionaries and the evaluation of translations and error analysis. The book is intended for teachers in translator-training institutions, but it can also be used by students for self-training.

Themes, syntax and other necessary steps in the network analysis of texts: a research paper

Recent approaches to the qualitative analysis of texts afford visual depictions of words as networks. Yet network characteristics can also be quantified, enabling one to draw probabilistic inferences about a population of texts from a sample of texts-encoded-as-networks. This article describes three types of ambiguity (and related methodological problems) that arise during three necessary steps in the quantification of texts as networks: idiomatic ambiguity (in the identification of themes [or nodes]); illocutionary ambiguity (in the identification of syntactic links [or arcs]); and relevance ambiguity (in the identification of network characteristics). As one moves from theme to syntax to network, not only does one add complexity to one's conclusions, but one also adds complexity to the encoding process as distinct types of linguistic ambiguity must be resolved. The added complexity of network encoding will be unnecessary for most research questions - questions that might better be addressed via thematic or semantic text analysis.

ETHNICITY, NATIONALISM, RACE, MINORITY: A SEMANTIC/ONOMANTIC EXERCISE (PART ONE)

All social scientists (including sociologists) face mounting ambiguity in their use of keywords because new meanings are continuously stipulated for them by scholars unwilling to coin new terms and face resulting criticisms. The established procedure for handling this situation is to apply semantic analysis of texts in order to identify the various important meanings of any given term, hoping that users will be able to keep them in mind when they find these words in their various theoretical and research contexts, and in the context of diverse discourse communities. The `onomantic' method, which seeks to add unambiguous terms to the ambiguous ones, provides a procedure for disambiguating texts: in addition to whatever terms a writer would naturally wish to use, it becomes possible to provide a supplementary term that can easily remove the ambiguity. In order to illustrate this procedure and, at the same time, to contribute to the development of the vocabulary for ethnicity research, concrete examples are offered. Five important meanings of ethnicity are identified, in a hierarchical relationship to each other: the generic concept; modern and pre-modern senses; the main pre-modern sense, and one of the modern senses. Unequivocal terms for each of these meanings of ethnicity are proposed. Overlapping terms like minority, race, and nationalism are also examined. The distinctions between numerical minority/majority; and politically dominant/marginalised are clarified so as to propose terms that can be used, for example, to refer to dominant minorities, marginalised minorities, and `non-ethnic minorities'. Similarly, the many overlapping senses of nationalism and race will be discussed and unequivocal terms suggested for each of them in Part Two, in the next issue.

Resolving pronoun references

Two approaches to the problem of resolving pronoun references are presented. The first is a naive algorithm that works by traversing the surface parse trees of the sentences of the text in a particular order looking for noun phrases of the correct gender and number. The algorithm clearly does not work in all cases, but the results of an examination of several hundred examples from published texts show that it performs remarkably well. In the second approach, it is shown how pronoun solution can be handled in a comprehensive system for semantic analysis of English texts. The system is described, and it is shown in a detailed treatment of several examples how semantic analysis locates the antecedents of most pronouns as a by-product. Included are the classic examples of Winograd and Charniak.

Pronoun resolution

Two approaches to the problem of pronoun resolution are presented. The first is a naive algorithm that works by traversing the surface parse trees of the sentences of the text in a particular order looking for noun phrases of the correct gender and number. The algorithm is shown to incorporate many, though not all, of the constraints on co-referentiality between a nonreflective pronoun and a possible antecedent, which have been discovered recently by linguists. The algorithm clearly does not work in all cases, but the results of an examination of several hundred examples from published texts show that it performs remarkably well.In the second approach, it is shown how pronoun resolution is handled in a comprehensive system for semantic analysis of English texts. The system consists of four basic semantic operations which work by accessing a data base of 'World knowledge" inferences, which are drawn selectively and in a context-dependent way in response to the operations. The first two operations seek to satisfy the demands made by predicates on the nature of their arguments and to discover the relations between sentences. The third operation - knitting - recognizes and merges redundant expressions. These three operations frequently result in a pronoun reference being resolved as a by-product. The fourth operation seeks to resolve those pronouns not resolved by the first three. It involves a bidirectional search of the text and 'World knowledge" for an appropriate chain of inference and utilizes the efficiency of the naive algorithm.Four examples, including the classic examples of Winograd and Charniak, are presented that demonstrate pronoun resolution within the semantic approach.

Semantic analysis of an ALGOL text

Semantic analysis of an ALGOL text

Semantic Analysis of Texts by Computer

1. Osgood1 has introduced a form of semantic analysis, the semantic differential, applicable to any arbitrarily chosen set of 'concepts' (usually but not necessarily nouns or noun phrases), in which pairs of adjective antonyms are used as measuring scales, by asking informants to indicate the degree to which each adjective or its opposite accurately describes the concept. It follows that any two concepts whose measurements on a large number of scales are very similar share something which may tentatively be called meaning. The whole matrix of such measurements may be factor-analyzed, by concepts or by scales, to yield clusters or dimensions of meaning, or, to use Osgood's expression, dimensions of semantic space. Weinreich2 has pointed out some theoretical objections to this procedure. First of all, the 'meaning' thus arrived at should not be confused in any way with dictionary meaning. It is affective, connotative; in other words, it is only that portion of meaning which comes through the rather fine sieve of forced metaphor. As such, it is of interest to the student of human cognition, but of less interest to the linguist or in particular the lexicographer. Second, many of the studies performed by Osgood seemed to imply the universality of the dimensions of semantic space. While this may conceivably be true (and important) for all human beings, it does not follow that all of semantic space is characterized by the same set of dimensions. That is, just as a physicist conceives of a space with local variations in curvature, we may conceive of a semantic space with local variations in complexity. Thus, within a certain limited semantic field, dimensions or contrasts may be in use which do not occur elsewhere (e.g. flat vs. plain). Some variation of this kind turns up in The measurement of meaning. Another example of this local variation is found in O'Dowd and Beardslee's work3 on occupational images, which shows the apparent bifurcation of one of Osgood's basic dimensions. Weinreich's suggestion that the game of twenty questions may provide a better semantic method than the universally-applied scales of the