Adaptive Network Based Fuzzy Inference System For Speech Recognition Through Subtractive Clustering

Silarbi, Samiya; Bendahmane, Abderrahmane; Benyettou, Abdelkader

doi:10.5121/ijaia.2014.5604

Cited by 8 publications

(5 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Bu ağı oluşturmak için beş katman kullanılır. Her katman, Şekil 1'de gösterilen birkaç düğüm yapısını içermektedir (Silarbi, Abderrahmane & Benyettou, 2014;Yakut & Süzülmüş, 2020).…”

Section: Yöntem 31 Anfis Yöntemi Ve Yapısıunclassified

Forecasting GDP in Türki̇ye Using Anfis Method

ŞENCAN¹,

Şahin²

2022

Kafkas Üniversitesi İktisadi Ve İdari Bilimler Fakültesi Dergisi

View full text Add to dashboard Cite

Gayri Safi Yurtiçi Hasıla (GSYİH), bir ülkenin makro ekonomik durumunun önemli göstergelerinden biridir. GSYİH büyümesinin büyük bir hassasiyetle tahmin edilmesi; politika yapıcılar, merkez bankaları ve diğer ekonomik aktörler için mali ve para politikalarının belirlenmesi ve planlanmasında önem taşımaktadır. Bu çalışmada, Türkiye’nin ekonomik büyümesini (GSYİH) tahmin etmek için yapay zekâ yöntemlerinden biri olan uyarlamalı ağ tabanlı bulanık mantık çıkarım sistemi (ANFIS) kullanılmıştır. İhracat, ithalat, devlet harcamaları, tüketici fiyat endeksi (TÜFE) ve enflasyon oranı girdilerine bağlı olarak ekonomik büyüme yani GSYİH tahmin edilmiştir. ANFIS modelinin güvenilirliği, çeşitli istatistiksel göstergeler kullanılarak belirlenmiştir. Ortalama mutlak yüzde hata oranı (MAPE) 1.5137 ve determinasyon katsayısı (R2) 0.9949 olarak elde edilmiştir. Bu değerlerin kabul edilebilir seviyelerde olduğu ve ekonomik büyümenin tahmin edilmesinde bu modelin başarıyla kullanılabileceği görülmüştür. Çalışmadan elde edilen sonuçlar, GSYİH büyümesini büyük bir hassasiyetle tahmin etmek, maliye ve ekonomi politikalarını belirlemek ve planlamak için kullanılabilir.

show abstract

Section: Yöntem 31 Anfis Yöntemi Ve Yapısıunclassified

Forecasting GDP in Türki̇ye Using Anfis Method

ŞENCAN¹,

Şahin²

2022

Kafkas Üniversitesi İktisadi Ve İdari Bilimler Fakültesi Dergisi

View full text Add to dashboard Cite

show abstract

“…The classical fuzzy set (crisp set) is used to define the high, mid-high, mid, mid-low, and low vowels that indicate the acceptable values of each word or utterance for the Al-Quran recitation from the GMM-HMM parameter estimation. Fuzzy correction mapping model in time series consists two main steps, which are structured identification and threshold parameter optimization [21]. For the structure identification, the number of membership functions and fuzzy if-then is applies to MFCC estimated parameters.…”

Section: Time Series Classification-threshold Processmentioning

confidence: 99%

A Model of Correction Mapping for Al-Quran Recitation Performance Evaluation Engine

Shafie¹

2019

IJATCSE

View full text Add to dashboard Cite

The main objective of the research is to develop the correction-mapping model for Al-Quran recitation performance evaluation engine. Machine learning and Digital Signal Processing techniques are applied in representing and analyzing the recitation speech signal. Consequently, a form of recitation correction results is derived and formulated for the final performance evaluation. The proposed corrective mapping model demonstrated in this paper confronted, but not limited to, with the challenging issues of variability of speaker recitation, recitation representation, speaker adaptation, feature extraction, parameters estimation and threshold process classification. The experimental results concluded the Al-Quran automated correction system knowns as Intelligent Quran Recitation Assistant (nur-IQRA) will be able to fulfil the current and future trends of digital society.

show abstract

“…For example, repository staff works to create quality records for their full-text documents, guaranteeing access to these items. However, a repository may also create metadata-only records for any faculty publication they cannot include in their collection, such as works published in journals with restrictive copyright policies [5]. The effectiveness of library collection services is an important element for a library itself because users can get the information needed by utilizing the collections provided by the library.…”

Section: Introductionmentioning

confidence: 99%

Design and Analysis of the E-Journal Repository for Final Projects and Student Practices in the D3 Computer Engineering

Afriliana

Bahar

Basit

2020

International Journal of Science, Technology & Management

View full text Add to dashboard Cite

Higher education institutions are the main institutions in creating young people who can be ready to enter the world of work and industry, and one of the educational programs that take part in vocational education. Vocational education is higher education that is directed at focusing certain applied skills. Harapan Bersama of Polytechnic is one of the vocational education institutions that provides Human Resources (HR) to become the younger generation as practitioners in industry, companies, or other stoke holders. In the D3 Computer Engineering Study Program, each batch can graduate approximately 300 graduates, where each graduate will produce one scientific work from the final project research conducted. Likewise, with students who do practical work (KP), every student who does KP will produce scientific papers. But all the results of this scientific work, namely the research report of the Final Project and the Job Training Report are still being submitted to the Study Program in one compact disk (CD). This turned out to be a problem for the Study Program because from each batch, around 300 compact disks (CDs) were collected from the results of scientific work from the practical work and 300 compact disks (CDs) from the final project research outputs. Physical data in the form of a compact disk (CD) requires a large space for its placement. Meanwhile, the condition of the D3 Computer Engineering study program room makes it impossible to place the physical form. And along with technological developments, this physical form can be replaced by soft files in the form of e-journals or e-reports. This needs to be done in line with the paradigm shift from conventional libraries to e-libraries. With the existence of SIPESTA or the information system for the completion of this Final Project, the results of writing scientific papers are uploaded and become a digital trace of all scientific papers that have been carried out by all D3 Computer Engineering students. And so if this data is needed it will be easy to find because the database has been well managed.

show abstract

Adaptive Network Based Fuzzy Inference System For Speech Recognition Through Subtractive Clustering

Abstract: ABSTRACT

Cited by 8 publications

References 28 publications

Forecasting GDP in Türki̇ye Using Anfis Method

Forecasting GDP in Türki̇ye Using Anfis Method

A Model of Correction Mapping for Al-Quran Recitation Performance Evaluation Engine

Design and Analysis of the E-Journal Repository for Final Projects and Student Practices in the D3 Computer Engineering

Contact Info

Product

Resources

About