The development of intuitive knowledge classifier and the modeling of domain dependent data

Kahraman, Hamdi Tolga; Sağıroğlu, Şeref; Çolak, İlhami

doi:10.1016/j.knosys.2012.08.009

Cited by 112 publications

(55 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…An interesting approach is presented in [11] for modeling the knowledge of e-Learning users under different domains. The approach is composed of a generic domain object model, user modeling, weight adjusting method, and a classification algorithm.…”

Section: Related Workmentioning

confidence: 99%

“…The proposed framework for the automatic assessment of knowledge levels tries to consolidate the efforts of several researchers and standards [11]- [13], [16]. In addition, it includes the idea of mobile learning [2], [6] and flexible e-Learning environments.…”

Section: The Proposed Frameworkmentioning

confidence: 99%

“…Further collections of some behaviors are gathered and measured in [11], then a classification scheme is used to evaluate the users knowledge.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Knowledge Level Assessment in e-Learning Systems Using Machine Learning and User Activity Analysis

Ghatasheh¹

2015

ijacsa

View full text Add to dashboard Cite

Abstract-Electronic Learning has been one of the foremost trends in education so far. Such importance draws the attention to an important shift in the educational paradigm. Due to the complexity of the evolving paradigm, the prospective dynamics of learning require an evolution of knowledge delivery and evaluation. This research work tries to put in hand a futuristic design of an autonomous and intelligent e-Learning system. In which machine learning and user activity analysis play the role of an automatic evaluator for the knowledge level. It is important to assess the knowledge level in order to adapt content presentation and to have more realistic evaluation of online learners. Several classification algorithms are applied to predict the knowledge level of the learners and the corresponding results are reported. Furthermore, this research proposes a modern design of a dynamic learning environment that goes along the most recent trends in e-Learning. The experimental results illustrate an overall performance superiority of a support vector machine model in evaluating the knowledge levels; having 98.6% of correctly classified instances with 0.0069 mean absolute error.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: The Proposed Frameworkmentioning

confidence: 99%

See 1 more Smart Citation

Knowledge Level Assessment in e-Learning Systems Using Machine Learning and User Activity Analysis

Ghatasheh¹

2015

ijacsa

View full text Add to dashboard Cite

show abstract

“…The Wine dataset was tested again in [14], but with slightly worse results. For section 5.4, a new classifier was tried in [13] for classifying knowledge about web page use. While their classifier was superior, they made comparisons with a Bayes and a k-NN classifier that did not perform better than the new model suggested here.…”

Section: Related Workmentioning

confidence: 99%

A single-pass classifier for categorical data

Greer

2017

IJCSYSE

View full text Add to dashboard Cite

-This paper describes a new method for classifying a dataset that partitions elements into their categories. It has relations with neural networks but a slightly different structure, requiring only a single pass through the classifier to generate the weight sets. A grid-like structure is required as part of a novel idea of converting a 1-D row of real values into a 2-D structure of value bands. Each cell in any band then stores a distinct set of weights, to represent its own importance and its relation to each output category. During classification, all of the output weight lists can be retrieved and summed to produce a probability for what the correct output category is. The bands possibly work like hidden layers of neurons, but they are variable specific, making the process orthogonal. The construction process can be a single update process without iterations, making it potentially much faster. It can also be compared with k-NN and may be practical for partial or competitive updating.

show abstract

“…Bu ölçüm ya da gözlemlerin her biri öznitelik (feature) olarak adlandırılır. Verilerden anlamlı bilgilerin çıkarılması sürecinde, yapılması gereken ilk işlem öznitelikler ile çıkış verilerinin birbiri ile olan bağımlılığının ortaya konulmasıdır [1]. Bunun ilk sebebi oluşturulacak model ister sınıflandırma (classification) ister bağlanım (regression) için kullanılacak olsun, hesaplama maliyeti girişteki öznitelik sayısına bağlıdır [2].…”

unclassified

Ağirliklandirilmiş Koşullu Karşilikli Bi̇lgi̇ İle Özni̇teli̇k Seçi̇mi̇

Çelik¹,

Bilge²

2015

Gazi Üniversitesi Mühendislik-Mimarlık Fakültesi Dergisi

View full text Add to dashboard Cite

ÖZETBüyük verilerin işlenmesi ve bu verilerden anlamlı bilgilerin çıkarılması veri madenciliğinin önemli konularından biridir. Uygulamada eldeki verilerin probleme uygun olup olmadığı bilinemez ve ilgisiz veriler oluşturulacak modelin karmaşıklığını arttırır. Daha basit ve düşük maliyetli modeller oluşturmak için problem parametreleri üzerinde boyut indirgeme yaklaşımları uygulanır. Bilgi teorisi tabanlı karşılıklı bilgi yaklaşımları boyut indirgemede yaygın olarak kullanılır. Bu yaklaşımlarda boyut indirgemesi yapılırken veri kümesinden elde edilecek alt kümede, girişlerin birbirleri ile minimum fazlalık ve çıkışla maksimum bağımlılığa sahip olması amaçlanmaktadır. Bu koşulu sağlamak için önerilen yaklaşımların özniteliklerin seçiminde kullandıkları sezgisel fonksiyonlar, bağımlılık ve fazlalık bilgileri arasındaki ilişkiyi problemden bağımsız ve sabit bir değişkenle kontrol eder. Bu çalışmada yeni bir karşılıklı bilgi yaklaşımı önerilmiştir. Önerilen yaklaşımda kullanılan sezgisel fonksiyon, fazlalığın seçim üzerindeki etkinliğini, özniteliklerin sınıfla olan karşılıklı bilgisi ve birbirleri ile olan karşılıklı bilgilerinin arasındaki ilişkiyi değerlendirerek ağırlıklandırır. Benzer şekilde maksimum bağımlılık için hem koşullu karşılıklı bilgi hem de karşılıklı bilgi hesaplanır. Böylelikle önerilen sezgisel fonksiyon, değişen problemler karşısında dinamik bir yaklaşım sergiler. Test sonuçları önerilen yaklaşımın başarısını göstermektedir.Anahtar Kelimeler: Öznitelik seçimi, koşullu karşılıklı bilgi, maksimum bağımlılık, minimum fazlalık FEATURE SELECTION WITH WEIGHTED CONDITIONAL MUTUAL INFORMATION ABSTRACTHuge data processing and extracting the meaningful information from those data is one of the important topics in data mining. In practice, it cannot be known whether present data are relevant to the problem, and irrelevant data increase the complexity of the prospective model. Dimensionality reduction approaches are applied to the problem parameters to build simpler and low cost models. Information theory based mutual information approaches are commonly used on dimensionality reduction. In these approaches, it is aimed to have the minimum redundancy and maximum dependency in outputs between inputs in the subset obtained from the data set in the execution of the size degradation. However the heuristic functions which are used in the proposed approaches to ensure this condition, control the relation between dependency and redundancy with the fixed parameter and problem-independent. In this study, a new mutual information approach is proposed. The heuristic function used in this proposed approach weights the effectiveness of redundancy on selection by evaluating the relationship between mutual information of features with class and mutual information of features among themselves. Similarly, both conditional mutual information and mutual information are calculated for maximum dependency. Thus the proposed heuristic function presents a dynamic approach to variety of problems. The obtained results of the tests poi...

show abstract

The development of intuitive knowledge classifier and the modeling of domain dependent data

Cited by 112 publications

References 22 publications

Knowledge Level Assessment in e-Learning Systems Using Machine Learning and User Activity Analysis

Knowledge Level Assessment in e-Learning Systems Using Machine Learning and User Activity Analysis

A single-pass classifier for categorical data

Ağirliklandirilmiş Koşullu Karşilikli Bi̇lgi̇ İle Özni̇teli̇k Seçi̇mi̇

Contact Info

Product

Resources

About