Separate Text Files Research Articles

Detection of K-complexes in EEG waveform images using faster R-CNN and deep transfer learning

BackgroundThe electroencephalography (EEG) signal carries important information about the electrical activity of the brain, which may reveal many pathologies. This information is carried in certain waveforms and events, one of which is the K-complex. It is used by neurologists to diagnose neurophysiologic and cognitive disorders as well as sleep studies.Existing detection methods largely depend on tedious, time-consuming, and error-prone manual inspection of the EEG waveform.MethodsIn this paper, a highly accurate K-complex detection system is developed. Based on multiple convolutional neural network (CNN) feature extraction backbones and EEG waveform images, a regions with faster regions with convolutional neural networks (Faster R-CNN) detector was designed, trained, and tested. Extensive performance evaluation was performed using four deep transfer learning feature extraction models (AlexNet, ResNet-101, VGG19 and Inceptionv3). The dataset was comprised of 10948 images of EEG waveforms, with the location of the K-complexes included as separate text files containing the bounding boxes information.ResultsThe Inceptionv3 and VGG19-based detectors performed consistently high (i.e., up to 99.8% precision and 0.2% miss rate) over different testing scenarios, in which the number of training images was varied from 60% to 80% and the positive overlap threshold was increased from 60% to 90%.ConclusionsOur automated method appears to be a highly accurate automatic K-complex detection in real-time that can aid practitioners in speedy EEG inspection.Graphical

Development of Automatic Rule-based Semantic Tagger and Karaka Analyzer for Hindi

Hindi is the third most-spoken language in the world (615 million speakers) and has the fourth highest native speakers (341 million). It is an inflectionally rich and relatively free word-order language with an immense vocabulary set. Despite being such a celebrated language across the globe, very few Natural Language Processing (NLP) applications and tools have been developed to support it computationally. Moreover, most of the existing ones are not efficient enough due to the lack of semantic information (or contextual knowledge). Hindi grammar is based on Paninian grammar and derives most of its rules from it. Paninian grammar very aggressively highlights the role of karaka theory in free-word order languages. In this article, we present an application that extracts all possible karakas from simple Hindi sentences with an accuracy of 84.2% and an F1 score of 88.5%. We consider features such as Parts of Speech tags, post-position markers (vibhaktis), semantic tags for nouns and syntactic structure to grab the context in different-sized word windows within a sentence. With the help of these features, we built a rule-based inference engine to extract karakas from a sentence. The application takes in a text file with clean (without punctuation) simple Hindi sentences and gives back karaka tagged sentences in a separate text file as output.

THE ALGORITHM OF CONSTRUCTING CONTROL FLOW GRAPH BASED ON A PROGRAM WRITTEN IN C

Actuality. The work considers the problem of automated construction of the control flow graph using the text of a program written in C. The graph construction is an important step in the structural testing of embedded systems. Objective. The work is aimed at creation of a fast algorithm for constructing the control flow graph from a source code. Such an automatically generated graph can be used by existing tools for automated testing. Method. We propose an algorithm for constructing the control flow graph from the program written in C language, which preprocesses the source code of the program by removing comments and blank lines, determines the number of vertices and edges of the control flow graph by means of program text syntactic analysis, and then formats, fills and stores the incidence matrix in a separate text file. This file can be passed as input data for some existing automated testing tools. Moreover, the content of the text file can be visualized by existing tools for graphic representation of graphs. Results. We’ve developed a software module that implements the proposed algorithm. The module can be used for experiments aimed at studying the dependence of the control flow graph’s time construction on the amount of lines in a source code. Conclusions. The conducted experiments confirmed the operability of the proposed algorithm and the software module developed on its basis. The results have proved the applicability of the algorithm, thus it can be recommended for further use together with automated testing tools to reduce the time required for testing of embedded software.

Design of a reference intelligence expert system for sheep breeding in national private farms

The paper deals with the design of a reference intelligence expert system for sheep breeding. The process of its design and its functional capabilities are described. The developed reference intelligence expert system makes it possible to display recommendations and tips on the computer screen to select a rational and effective sheep breeding technology, as well as to print the received data. The developed system has a two-tier structure, namely, dialog selection of input criteria (data), as well as a module for logical processing and output of reasonable technology recommendations (output data). The step-by-step method of designing the system is determined. The first step is to obtain a technical specification from the customer (farm, enterprise, farmer, etc.) for the development of a reference system. The second step is to determine criteria for the technology in keeping with the farm requirement specification according to the literature and sheep breeding requirements. The third step is to identify the most important factors influencing the process of sheep raising for each of these criteria. The fourth step is to define the main production rules for which the system will be programmed, namely, the module for logical knowledge processing for this technology (and these are the input parameters (factors) when designing the system). The fifth step is to determine the main output criteria (factors) that will be displayed after processing the input rules of the system based on logical deduction rules according to the appropriate sheep breeding technology. The sixth step is to design buttons for more convenient system usage, in addition, if necessary, or at the request of the customer (for example, a button for deleting previous information in windows, a button for storing recommendations in a separate text file, the exit button, etc.). The seventh step is to test the system by users and the customer. The eighth step is to adopt the developed system by the customer and correct it, if necessary. The ninth step is to provide maintenance of the developed system. The block of input data of the developed intelligence expert system has the following elements: a) sheep handling (pasture-stall, stall-pasture); b) sheep condition (higher condition, average condition, lower than average condition); c) sheep breeding type (meat, milk, meat and milk, wool); d) sheep feeding (meat-and-wool sheep, rams of meat-and-goat sheep, sucking females to lactation); e) methods of breeding (purebred, crossing). The block of output factors where appropriate rational recommendations for sheep breeding technology comprises: a) recommendations for sheep handling; b) recommendations for treatment of sheep diseases; c) recommendations for sheep feeding; d) sheep slaughter and storage of products; e) recommended sheep breeds; f) recommendations for sheep breeding; g) characteristics of products. The developed system is intended primarily for private national farmers or interested private householders. It is noted that the use of this system will allow private households to increase the production and quality of meat, wool, lambskin and furs when breeding sheep. This will reduce the cost of the breeding technology, save the farmer time, allow to save on medicines, feeds, which, in turn, will increase profits and profitability of farms.

Exploring trends in the codes of ethics of the Fortune 100 and Global 100 corporations

PurposeThe purpose of this paper is to explore emerging trends in the content of codes of ethics of US Fortune 100 and Global 100 corporate web sites through a longitudinal study.Design/methodology/approachThe web sites of the 200 companies were surfed and the relevant documents of the codes of ethics were extracted to separate text files. The computer files were then mined using the customized developed software and each ethical keyword was counted. The number of filed codes of ethics was 95 for both the 100 US‐based and the 100 Global‐based companies.FindingsIn addition to the content analysis of the ethic codes of the studied groups and finding high‐frequency ethical keywords, the results of this study indicate a convergence of the contents. Moreover, the results also show that the ethical codes are now more readily available on corporations’ web sites; that is a sign of higher level of disclosure of ethical codes compared to that in 2006. Finally, this research proposes some hypotheses to explain the changes from 2006 to 2009.Practical implicationsMany smaller corporations and start‐up companies can benefit from the results of this study by comparing their codes of ethics with those of the major US and global companies, using key ethical phrases that are discussed here. Moreover, US companies wanting to establish new subsidiaries in other countries can also take advantage of the results of this study. They can find out what are the common dissimilarities between American corporates and other international firms, from ethical point of view, when they want to reach a uniform code of ethics.Social implicationsLongitudinal study of the content of codes of ethics can help in encouraging firms to give consideration to ethical issues. This research is the first attempt to study the ethical standards adopted by the major US and global corporations, before and after the global financial crisis.Originality/valueThis study analyses the content of codes of ethics of the world's top firms and compares Fortune 500 and global 500 companies, considering the frequency of ethical keywords on their codes. It also compares the similarities and differences and indicates whether the content is divergent or convergent. The study also shows how the disclosure of codes of ethics has changed in the time‐span of research.

Creating a Multi-axis Machining Postprocessor

This paper focuses on the postprocessor creation process. When using standard commercially available postprocessors it is often very difficult to modify its internal source code, and it is a very complex process, in many cases even impossible, to implement the newly-developed functions. It is therefore very important to have a method for creating a postprocessor for any CAM system, which allows CL data (Cutter Location data) to be generated to a separate text file. The goal of our work is to verify the proposed method for creating a postprocessor. Postprocessor functions for multi-axis machiningare dealt with in this work. A file with CL data must be translated by the postprocessor into an NC program that has been customized for a specific production machine and its control system. The postprocessor is therefore verified by applications for machining free-form surfaces of complex parts, and by executing the NC programs that are generated on real machine tools. This is also presented here.

A Bayesian Approach for the Categorization of Radiology Reports

We sought to develop a Bayesian-filter that could distinguish positive radiology computed tomography (CT) reports of appendicitis from negative reports with no appendicitis. Standard unstructured electronic text radiology reports containing the key word appendicitis were obtained using a Java-based text search engine from a hospital General Electric PACS system. A total of 500 randomly selected reports from multiple radiologists were then manually categorized and merged into two separate text files: 250 positive reports and 250 negative findings of appendicitis. The two text files were then processed by the freely available UNIX-based software dbacl 1.9, a digramic Bayesian classifier for text recognition, on a Linux based Pentium 4 system. The software was then trained on the two separate merged text files categories of positive and negative appendicitis. The ability of the Bayesian filter to discriminate between reports of negative and positive appendicitis images was then tested on 100 randomly selected reports of appendicitis: 50 positive cases and 50 negative cases. The training time for the Bayesian filter was approximately 2 seconds. The Bayesian filter subsequently was able to categorize 50 of 50 positive reports of appendicitis and 50 of 50 reports of negative appendicitis, in less than 10 seconds. A Bayesian-filter system can be used to quickly categorize radiology report findings and automatically determine after training, with a high degree of accuracy, whether the reports have text findings of a specific diagnosis. The Bayesian filter can potentially be applied to any type of radiologic report finding and any relevant category.

SEPARATE.BAS: A microcomputer program to assist in the textual analysis of qualitative data

Using microcomputer technology for processing textual data offers increased capabilities for the qualitative researcher in speech communication. This article presents an original microcomputer program, SEPARATE.BAS. This program assists the researcher in dividing transcripts of interviews or discussions into separate text files for each participant. The article discusses several applications for the program in qualitative research.