Automatic Documentation and Mathematical Linguistics

This paper presents a software complex that provides the interaction of software systems to solve the problem of integrating the attribute data of information objects, which is based on the developed and tested mathematical and programming software. This complex can be used to increase the efficiency of fuzzy record linkage processes in data coordination in information resources, systems, and databases.

This paper discusses the possibility of expanding the ideas of the validity of medical decisions of a diagnostic nature, which are made in the framework of so-called evidence-based medicine. An approach is proposed that allows building special data in the process of intelligent analysis of accumulated empirical data, which characterize the causality of a diagnosed effect–logical conditions (characteristic functions) that take the value true in all instances of the presence of the target effect and the value false for all instances of its absence in the training sample of precedents. This problem is solved based on the expanding sequences of training samples using: (a) a formal refinement of the concept of similarity of precedent descriptions as a binary algebraic operation, and (b) a mathematical technique for generating empirical dependences in the style of the JSM method of automated support for scientific research. The features and capabilities of the developed approach are described based on the example of solving the problem of analyzing the causes and predicting the pseudoprogression of brain tumors.

This paper describes the possibility of establishing the semantic proximity of scientific texts by the method of their automatic classification based on the compression of annotations. The idea of the method is that the compression algorithms such as PPM (prediction by partial matching) compress terminologically similar texts much better than distant ones. If a kernel of publications (an analogue of a training set) is formed for each classified topic, then the best proportion of compression will indicate that the classified text belongs to the corresponding topic. Thirty thematic categories were determined; for each of them, annotations of approximately 500 publications were received in the Scopus database, out of which 100 annotations for the kernel and 20 annotations for testing were selected in different ways. It was found that building a kernel based on highly cited publications revealed an error level of up to 12 against 32% in the case of random sampling. The quality of classification is also affected by the initial number of categories: the fewer the categories that participate in the classification and the more terminological differences exist between them, the higher its quality is.