All issues

In general, solving the information explosion problem can be delegated to systems for automatic processing of digital data. These systems are intended for recognizing, sorting, meaningfully processing and presenting data in formats readable and interpretable by humans. The creation of intelligent knowledge extraction systems that handle unstructured data would be a natural solution in this area. At the same time, the evident progress in these tasks for structured data contrasts with the limited success of unstructured data processing, and, in particular, document processing. Currently, this research area is undergoing active development and investigation. The present paper is a systematic survey on both Russian and international publications that are dedicated to the leading trend in automatic text data processing: Text Mining (TM). We cover the main tasks and notions of TM, as well as its place in the current AI landscape. Furthermore, we analyze the complications that arise during the processing of texts written in natural language (NLP) which are weakly structured and often provide ambiguous linguistic information. We describe the stages of text data preparation, cleaning, and selecting features which, alongside the data obtained via morphological, syntactic, and semantic analysis, constitute the input for the TM process. This process can be represented as mapping a set of text documents to «knowledge». Using the case of stock trading, we demonstrate the formalization of the problem of making a trade decision based on a set of analytical recommendations. Examples of such mappings are methods of Information Retrieval (IR), text summarization, sentiment analysis, document classification and clustering, etc. The common point of all tasks and techniques of TM is the selection of word forms and their derivatives used to recognize content in NL symbol sequences. Considering IR as an example, we examine classic types of search, such as searching for word forms, phrases, patterns and concepts. Additionally, we consider the augmentation of patterns with syntactic and semantic information. Next, we provide a general description of all NLP instruments: morphological, syntactic, semantic and pragmatic analysis. Finally, we end the paper with a comparative analysis of modern TM tools which can be helpful for selecting a suitable TM platform based on the user’s needs and skills.

The technology of creating patterns from natural language words (concepts) based on text data in the bag of words model is considered. Patterns are used to reduce the dimension of the original space in the description of documents and search for semantically related words by topic. The process of dimensionality reduction is implemented through the formation of patterns of latent features. The variety of structures of document relations is investigated in order to divide them into themes in the latent space.

It is considered that a given set of documents (objects) is divided into two non-overlapping classes, for the analysis of which it is necessary to use a common dictionary. The belonging of words to a common vocabulary is initially unknown. Class objects are considered as opposition to each other. Quantitative parameters of oppositionality are determined through the values of the stability of each feature and generalized assessments of objects according to non-overlapping sets of features.

To calculate the stability, the feature values are divided into non-intersecting intervals, the optimal boundaries of which are determined by a special criterion. The maximum stability is achieved under the condition that the boundaries of each interval contain values of one of the two classes.

The composition of features in sets (patterns of words) is formed from a sequence ordered by stability values. The process of formation of patterns and latent features based on them is implemented according to the rules of hierarchical agglomerative grouping.

A set of latent features is used for cluster analysis of documents using metric grouping algorithms. The analysis applies the coefficient of content authenticity based on the data on the belonging of documents to classes. The coefficient is a numerical characteristic of the dominance of class representatives in groups.

To divide documents into topics, it is proposed to use the union of groups in relation to their centers. As patterns for each topic, a sequence of words ordered by frequency of occurrence from a common dictionary is considered.

The results of a computational experiment on collections of abstracts of scientific dissertations are presented. Sequences of words from the general dictionary on 4 topics are formed.

We consider a portfolio with call option and the corresponding underlying asset under the standard assumption that stock-market price represents a random variable with lognormal distribution. Minimizing the variance hedging risk of the portfolio on the date of maturity of the call option we find a fraction of the asset per unit call option. As a direct consequence we derive the statistically fair lookback call option price in explicit form. In contrast to the famous Black–Scholes theory, any portfolio cannot be regarded as  risk-free because no additional transactions are supposed to be conducted over the life of the contract, but the sequence of independent portfolios will reduce risk to zero asymptotically. This property is illustrated in the experimental section using a dataset of daily stock prices of 37 leading US-based companies for the period from April 2006 to January 2013.

The method for mapping of intermediate representations (IR) set of C, C++ programs to vector embedding space is considered to create an empirical estimation framework for static performance prediction using LLVM compiler infrastructure. The usage of embeddings makes programs easier to compare due to avoiding Control Flow Graphs (CFG) and Data Flow Graphs (DFG) direct comparison. This method is based on transformation series of the initial IR such as: instrumentation — injection of artificial instructions in an instrumentation compiler’s pass depending on load offset delta in the current instruction compared to the previous one, mapping of instrumented IR into multidimensional vector with IR2Vec and dimension reduction with t-SNE (t-distributed stochastic neighbor embedding) method. The D1 cache miss ratio measured with perf stat tool is considered as performance metric. A heuristic criterion of programs having more or less cache miss ratio is given. This criterion is based on embeddings of programs in 2D-space. The instrumentation compiler’s pass developed in this work is described: how it generates and injects artificial instructions into IR within the used memory model. The software pipeline that implements the performance estimation based on LLVM compiler infrastructure is given. Computational experiments are performed on synthetic tests which are the sets of programs with the same CFGs but with different sequences of offsets used when accessing the one-dimensional array of a given size. The correlation coefficient between performance metric and distance to the worst program’s embedding is measured and proved to be negative regardless of t-SNE initialization. This fact proves the heuristic criterion to be true. The process of such synthetic tests generation is also considered. Moreover, the variety of performance metric in programs set in such a test is proposed as a metric to be improved with exploration of more tests generators.