Back-Propagation and K-Means Algorithms Comparison

Skorpil, Vladislav; Šťastný, Jiří

doi:10.1109/icosp.2006.345838

Cited by 24 publications

(15 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multi-layer Perceptron [2][3][4], Hidden Markov Models [5][6][7], Recurrent Neural Network [8,9] and Dynamic Time Warping [10][11][12] are some common methods applied to recognize the speech signal.…”

Section: Introductionmentioning

confidence: 99%

Measuring the performance of isolated spoken Malay speech recognition using Multi-layer Neural Networks

Seman

Bakar

2010

2010 International Conference on Science and Social Research (CSSR 2010)

View full text Add to dashboard Cite

Abstract-This paper describes speech signal modeling techniques which are suited to high performance and robust isolated word recognition. In this study, a speech recognition system is presented, specifically an isolated spoken Malay word recognizer which uses spontaneous and formally speeches collected from Parliament of Malaysia. Currently the vocabulary is limited to 25 words that can be pronounced exactly as it written and controls the distribution of the vocalic segments. The speech segmentation task is achieved by adopted energy based parameter and zero crossing rate measure with modification to better locates the beginning and ending points of speech from the spoken words. The training and recognition processes are realized by using Multi-layer Perceptron (MLP) Neural Networks with two-layer network configurations that are trained with stochastic error back-propagation to adjust its weights and biases after presentation of every training data. The Melfrequency Cepstral Coefficients (MFCCs) has been chosen as speech extraction approach from each segmented utterance as characteristic features for the word recognizer. Recognition results showed that the performance of the two-layer networks increased as the numbers of hidden neurons increased. The best network structures average classification rate is 84.731% with (150-25) configuration. Implementation results also showed that the conjugate gradient (CG) algorithm was more accurate and reliable than the Levenberg-Marquardt (LM) algorithm for the network complexities and data size considered in this study.

show abstract

Section: Introductionmentioning

confidence: 99%

Measuring the performance of isolated spoken Malay speech recognition using Multi-layer Neural Networks

Seman

Bakar

2010

2010 International Conference on Science and Social Research (CSSR 2010)

View full text Add to dashboard Cite

show abstract

“…This follows up on the exemplary neural network or k-means classifi cation (Škorpil, Šťastný, 2006;Trenz, Konečný, 2010 …”

Section: Discussionmentioning

confidence: 99%

Neural Network for Determining Risk Rate of Post-Heart Stroke Patients

Trenz¹,

Sepši²,

Konečný³

2014

Acta Univ. Agric. Silvic. Mendelianae Brun.

View full text Add to dashboard Cite

The ischemic heart disease presents an important health problem that aff ects a great part of the population and is the cause of one third of all deaths in the Czech Republic. The availability of data describing the patients' prognosis enables their further analysis, with the aim of lowering the patients' risk, by proposing optimum treatment. The main reason for creating the neural network model is not only to automate the process of establishing the risk rate of patients suff ering from ischemic heart disease, but also to adapt it for practical use in clinical conditions. Our aim is to identify especially the specifi c group of risk-rate patients whose well-timed preventive care can improve the quality and prolong the length of their lives. The aim of the paper is to propose a patient-parameter structure, using which we could create a suitable model based on a self-taught neural network. The emphasis is placed on identifying key descriptive parameters (in the form of a reduction of the available descriptive parameters) that are crucial for identifying the required patients, and simultaneously to achieve a portability of the model among individual clinical workplaces (availability of parameters).

show abstract

“…These models vary according to their complexity and effectivity. The most basic models propose that all indicators and dimensions have a similar weight in assessing the enterprises performance [35], whereas more complicated ones require several statistical or mathematical analyses to be performed [8,36,37], or even artificial neural networks engagement [38]. For the construction of the evaluation model, economic value added (EVA), data envelopment analysis (DEA) and sustainability value added (SVA) are used.…”

Section: Efficiency and Sustainability Assessment Modelsmentioning

confidence: 99%

Automated Sustainability Assessment System for Small and Medium Enterprises Reporting

Kassem

Trenz

2020

Sustainability

View full text Add to dashboard Cite

Sustainability assessment is a mainstream business activity that demonstrates the link between the organization’s strategy for and commitment to a sustainable global economy, and the prevention of economic crises. Small- and medium-sized businesses/enterprises (SMBs/SMEs) have significant effects on the European economy. However, because of a lot of restriction factors, like business risk, the high expenses of data collection and management, and the lack of resources, sustainability reporting is considered a superfluous and burdensome activity for them. The aim of this research is to propose an automated, comprehensive and simplified system for the sustainability assessment of SMEs. This system is achieved by implementing three main phases. The first phase includes key performance indicators design, which starts with the identification of various key performance indicators for comprehensive sustainability assessment, and ends with proposing an optimal set of KPIs (Key Performance Indicators) that can encompass long-term issues and be applicable to SMEs in the EU. The second phase involves a new comprehensive method of sustainability assessment for all KPIs designed in the first phase. Therefore, a multi-criteria model, which involves four main pillars of sustainability assessment (economic, environmental, social and governance), is proposed. It gives different enterprises the ability to verify and compare their efficiency and sustainability with other companies within the same sector in an almost automated manner. In the final phase, a simple and an automated information system (WEBRIS), which provides a suitable environment for SME sustainability reporting, is developed. Finally, this system is verified in a case study of the Czech breweries sector.

show abstract

Back-Propagation and K-Means Algorithms Comparison

Cited by 24 publications

References 2 publications

Measuring the performance of isolated spoken Malay speech recognition using Multi-layer Neural Networks

Measuring the performance of isolated spoken Malay speech recognition using Multi-layer Neural Networks

Neural Network for Determining Risk Rate of Post-Heart Stroke Patients

Automated Sustainability Assessment System for Small and Medium Enterprises Reporting

Contact Info

Product

Resources

About