Wyniki wyszukiwania - BazTech

1

Application of artificial neural networks in preliminary selection of pigmented lesions for further melanoma diagnosis

Kisielińska-Ptasznik Anna, Figielska Ewa

Zeszyty Naukowe Warszawskiej Wyższej Szkoły Informatyki

|

2020

|

nr 22

23--38

EN

The paper deals with the problem of preliminary selection of pigmented lesionsfor further melanoma diagnosis. Several algorithms for input data pre-processingare proposed and artificial neural network for the examination of pigmented lesions is used. Computational results are reported.

2

Analysis of data pre-processing methods for sentiment analysis of reviews

Parlar Tuba, Ozel Selma, Song Fei

Computer Science

|

2019

|

Vol. 20 (1)

123--141

EN

The goals of this study are to analyze the effects of data pre-processing methods for sentiment analysis and determine which of these pre-processing methods (and their combinations) are effective for English as well as for an agglutinative language like Turkish. We also try to answer the research question of whether there are any differences between agglutinative and non-agglutinative languages in terms of pre-processing methods for sentiment analysis. We find that the performance results for the English reviews are generally higher than those for the Turkish reviews due to the differences between the two languages in terms of vocabularies, writing styles, and agglutinative property of the Turkish language.

3

Data preprocessing and its influence on the adsorption capacity of calculated shale. Comparison of selected methods: A case study

Waszczuk P.

Mineralogia

|

2017

|

Vol. 48, iss. 1/4

157--165

EN

This article discusses the impact of choosing a data preprocessing method for the calculated gas density necessary to determine the sorption capacity of a material. The sample of gas-bearing shale was subjected to a volumetric sorption test. The obtained data, pressure and temperature were preprocessed by three methods: moving average, polynomial regression and locally weighted scatterplot smoothing. The results include the excess and absolute sorption calculated from data that were filtered, and data without pre-treatment and Langmuir isotherms’ coefficients for every case.

4

Some methods of pre-processing input data for neural networks

Kuźniar K., Zając M.

Computer Assisted Methods in Engineering and Science

|

2015

|

Vol. 22, no. 2

141--151

EN

Two techniques of data pre-processing for neural networks are considered in this paper: (i) data compression with the application of the principal component analysis method, and (ii) various forms of data scaling. The novelty of this paper is associated with compressed input data scaling by the rotation (by the “stretching”) in neural network. This approach can be treated as the new proposition for data preprocessing techniques. The influence of various types of input data pre-processing on the accuracy of neural network results is discussed by using numerical examples for the cases of natural frequency predictions of horizontal vibrations of load-bearing walls. It is concluded that a significant reduction in the neural network prediction errors is possible by conducting the appropriate input data transformation.

5

Dyskretyzacja danych numerycznych metodami przekształceń boolowskich

Jankowski C., Borowik G., Kowalski K.

Przegląd Telekomunikacyjny + Wiadomości Telekomunikacyjne

|

2014

|

nr 10

1334--1342

PL

Dyskretyzacja jest jednym z podstawowych zabiegów wstępnego przetwarzania tablic decyzyjnych. Przekształcenie ciągłych wartości atrybutów na ich dyskretne odpowiedniki umożliwia dalszą analizę za pomocą metod eksploracji danych. Od jakości dyskretyzacji zależy zatem dokładność przewidywań, uzyskanych za pomocą wyznaczania reguł decyzyjnych. Przedstawiono opis metody dyskretyzacji danych numerycznych w tablicach decyzyjnych metodami przekształceń boolowskich. Pokazano, iż użycie algorytmów, wywodzących się z syntezy logicznej, umożliwia uzyskanie dobrej jakościowo dyskretyzacji.

EN

Discretization is one of the most important parts of decision tables preprocessing. Transformation continuous values of attributes into discrete intervals allows further analysis using data mining methods. The accuracy of generated rules predictions relies on the quality of discretization. The paper contains a description of the method of discretization of numerical data in decision tables using boolean transformations. Has been shown that the use of algorithms derived from logic synthesis results in a good quality discretization.

6

Efficiency of linear and non-linear classifiers for gas identification from electrocatalytic gas sensor

Kalinowski P., Woźniak Ł., Strzelczyk A., Jasinski P., Jasinski G.

Metrology and Measurement Systems

|

2013

|

Vol. 20, nr 3

501--512

EN

Electrocatalytic gas sensors belong to the family of electrochemical solid state sensors. Their responses are acquired in the form of I-V plots as a result of application of cyclic voltammetry technique. In order to obtain information about the type of measured gas the multivariate data analysis and pattern classification techniques can be employed. However, there is a lack of information in literature about application of such techniques in case of standalone chemical sensors which are able to recognize more than one volatile compound. In this article we present the results of application of these techniques to the determination from a single electrocatalytic gas sensor of single concentrations of nitrogen dioxide, ammonia, sulfur dioxide and hydrogen sulfide. Two types of classifiers were evaluated, i.e. linear Partial Least Squares Discriminant Analysis (PLS-DA) and nonlinear Support Vector Machine (SVM). The efficiency of using PLS-DA and SVM methods are shown on both the raw voltammetric sensor responses and pre-processed responses using normalization and auto-scaling.

7

Neural networks for the analysis of mine-induced building vibrations

Kuźniar K.

Computer Assisted Mechanics and Engineering Sciences

|

2011

|

Vol. 18, no. 3

147-159

EN

A study of the capabilities of arti?cial neural networks in respect of selected problems of the analysis of mine-induced building vibrations is presented. Neural network technique was used for the prediction of building fundamental natural period, mapping of mining tremors parameters into response spectra from ground vibrations, soil-structure interaction analysis, simulation of building response to seismictype excitation. On the basis of the experimental data obtained from the measurements of kinematic excitations and dynamic responses of actual structures, training and testing patterns of neural networks were formulated. The obtained results lead to a conclusion that the neural technique gives possibility of e?cient, accurate enough for engineering, analysis of structural dynamics problems related to mineinduced excitations.

8

Wstępne przetwarzanie danych rzeczywistych na przykładzie analizy rezygnacji

Sulikowski P.

Metody Informatyki Stosowanej

|

2009

|

nr 1 (18)

99-106

EN

Customer churn is a grave problem for contemporary businesses. Rough sets method can be applied to build a model of the phenomenon. While analysing real-world data obtained directly from an interested company, the need for data pre-processing arises. For that reason, this paper introduces a procedure combining certain useful analytical activities. They can help improve credibility and interpretability of the final results.

9

Data mining-generation and visualisation of decision trees

Kwaśnicka H., Doczekalski M.

Systems Science

|

2002

|

Vol. 28, nr 3

63-84

EN

A computer system presented in the paper is developed as a data mining tool-it allows using large databases as a source for the process of decision tree generation and visualisation. The designed system (DTB&V-Decision Tree Builder and Visualiser) is able to perform data preprocessing, generation of decision trees followed by their post-processing and visualisation. DTB&V was tested using a number of databases commonly employed for such tasks.