A Study on Naïve Bayes Classifier Based Document Classification Scheme with an Apriori Feature Extraction

Jong-Yeol Yoo,Min-Ho Lee,Grace Aloyce,Dong-Min Yang 보안공학연구지원센터 2016 International Journal of Database Theory and Appli Vol.9 No.7

A document classifier is an essential tool for classifying the various types of documents being generated in the Big Data era. In recent years, the wide variety of information services available for use with smartphones and portable mobile devices (tablets) have provided a technique that efficiently classifies the quality of sorted data. A common type of document classification scheme is the naïve Bayes classifier. The Naïve Bayes scheme is based on performance classification, which varies widely depending on the method of extraction used in the document. In this paper, we propose a system model that offers feature extraction methods which combine frequency with associated words. This model is then applied to the Naïve Bayes classifier to precisely classify documents. This method is proposed as an alternative to using traditional classification techniques. In addition, experiments will be evaluated by the existing document classification techniques and the proposed techniques.

Detection of Rice Disease Using Bayes' Classifier and Minimum Distance Classifier

Sharma, Vikas,Mir, Aftab Ahmad,Sarwr, Abid Korea Multimedia Society 2020 The journal of multimedia information system Vol.7 No.1

Rice (Oryza Sativa) is an important source of food for the people of our country, even though of world also .It is also considered as the staple food of our country and we know agriculture is the main source country's economy, hence the crop of Rice plays a vital role over it. For increasing the growth and production of rice crop, ground-breaking technique for the detection of any type of disease occurring in rice can be detected and categorization of rice crop diseases has been proposed in this paper. In this research paper, we perform comparison between two classifiers namely MDC and Bayes' classifiers Survey over different digital image processing techniques has been done for the detection of disease in rice crops. The proposed technique involves the samples of 200 digital images of diseased rice leaf images of five different types of rice crop diseases. The overall accuracy that we achieved by using Bayes' Classifiers and MDC are 69.358 percent and 81.06 percent respectively.

Mutual Information 이론을 활용한 적정 변수선택법 적용 연구

이선미 대한의료정보학회 2005 Healthcare Informatics Research Vol.11 No.3

Objective: The purpose of this study was to explore the usability of a feature selection method based on the mutual information theory to increase predictive performance of a classifier in data mining. Methods: The HIV Cost and Services Utilization Study(HCSUS) dataset was used to apply the feature selection method to a classifier. Its contribution to increasing the predictive performance of the classifier was evaluated by comparing the Naive Bayes(NB) and the Logistic Regression(LG) models using different variables. The infrequent office visit representing limited health service utilization was selected as an outcome variable. HUGIN ResearcherTM 6.3 was used to train and test the NB models and SAS 8.0 was used for the LG modeling. Results: Higher AUC in the NB model was obtained using the variables selected by the mutual information based feature selection method(AUC=.639, CI=.611, .660); lower AUC using the variables defined by a previous study(AUC=.599, CI=.570, .620). There was no difference between the LG models with different variables. Conclusion: This study demonstrated the mutual information method may be useful in identifying relevant predictors as the feature selection method, which can contribute to an increase in the predictive performance of a classifier. (Journal of Korean Society of Medical Informatics 11-3,247-253, 2005)

Optimization of the Smoothing Parameter of the Adaptive Kernel Estimator used in Bayes Classifier - Application to Microarray Data Analysis

Yissam Lakhdar,El Hassan Sbai 보안공학연구지원센터 2015 International Journal of Software Engineering and Vol.9 No.3

In this work, we focus on nonparametric kernel methods for estimating the probability density function (pdf). The convergence of a kernel estimator depends crucially on the choice of the smoothing parameter. We present in this paper, a new method for optimizing the bandwidth of an estimator of the probability density function: the adaptive kernel estimator. This optimized estimator is used to construct the Bayes classifier. In this sense, we have proposed a new approach to optimize the pdf based on the statistical properties of the probability distributions of random variables. We adopt the maximum entropy principle (MEP) in order to determine the optimal value of the smoothing parameter used in the estimator. In the proposed criterion, the estimated probability density function is called optimal in the sense of having a minimum error rate of classifying data. Finally, we illustrate the robustness of our optimization process of the kernel estimation methods by using a set of DNA microarray data showing that our approach effectively improves the performance of the classification process.

Affective Classification Using Bayesian Classifier and Supervised Learning

Seong Youb Chung,Hyun Joong Yoon 제어로봇시스템학회 2012 제어로봇시스템학회 국제학술대회 논문집 Vol.2012 No.10

An affective classification technology plays a key role in the affective human and computer interaction. This paper presents an affective classification method based on the Bayes classifier and the supervisory learning. We newly define a weighted-log-posterior function for the Bayes classifier, instead of the posterior function or the likelihood function that is used in the ordinary Bayes classifier. The weighted-log-posterior function is represented as the weighted sum of likelihood function of each feature plus bias factor under the assumption of feature independence. The Bayes classifier finds an affective state with the maximum value of the weighted-log-posterior function. The weights and the bias factors are iteratively computed by using supervisory learning approach. In the implementation, the affective states are divided into two and three classes in valence dimension and arousal dimension, respectively. An open database for emotion analysis using electroencephalogram (DEAP) is used to evaluate the proposed method. The accuracies for valence and arousal classification are 66.6 % and 66.4 % for two classes and 53.4 % and 51.0 % for three classes, respectively.

온라인 쇼핑몰에서의 상담 대응 효율 개선을 위한 AI기반 상담 분류 기법

이건수(Keonsoo Lee),김중연(Jung-Yeon Kim),강병권(Byeong-Gwon Kang) 한국정보기술학회 2021 한국정보기술학회논문지 Vol.19 No.8

Online shopping has various advantages over offline shopping. In order to utilize such potential, it is important to realize customers’ intention in the process of shopping. Once the request is categorized, a proper agent, who responds to the request, can be easily assigned. In this paper, we propose a method of categorizing types of customers’ counseling requests asked by active or potential users of an online shopping website. The proposed method is composed of two sub-classifiers which are naive Bayes classifier and Gated Recurrent Unit(GRU) based classifier. These two sub-classifiers are trained using the document-term matrix and word embeddings generated from a dataset collected from an online shopping website. Predictions of these two classifiers are combined using ensemble method. We obtained 87.05% and 88.94% accuracy for naive Bayes and GRU based classifier, respectively, and the performance was improved by ensembling the two weak classifiers achieving up to 95.29% accuracy.

다중 레이블 나이브 베이즈 분류를 위한 새로운 사후확률 추정 방법에 관한 연구

김해천(Hae-Cheon Kim),이재성(Jaesung Lee) 한국컴퓨터정보학회 2018 韓國컴퓨터情報學會論文誌 Vol.23 No.6

A multi-label classification is to find multiple labels associated with the input pattern. Multi-label classification can be achieved by extending conventional single-label classification. Common extension techniques are known as Binary relevance, Label powerset, and Classifier chains. However, most of the extended multi-label naive bayes classifier has not been able to accurately estimate posterior probabilities because it does not reflect the label dependency. And the remaining extended multi-label naive bayes classifier has a problem that it is unstable to estimate posterior probability according to the label selection order. To estimate posterior probability well, we propose a new posterior probability estimation method that reflects the probability between all labels and labels efficiently. The proposed method reflects the correlation between labels. And we have confirmed through experiments that the extended multi-label naive bayes classifier using the proposed method has higher accuracy then the existing multi-label naive bayes classifiers.

포섭구조 일대다 지지벡터기계와 Naive Bayes 분류기를 이용한 효과적인 지문분류

홍진혁(Jin-Hyuk Hong),민준기(Jun-Ki Min),조웅근(Ung-Keun Cho),조성배(Sung-Bae Cho) 한국정보과학회 2006 정보과학회논문지 : 소프트웨어 및 응용 Vol.33 No.10

지문분류는 사전에 정의된 클래스로 입력된 지문을 분류하여 자동지문인식 시스템에서 비교해야할 지문의 수를 줄여준다. 지지벡터기계(support vector machine; SVM)는 패턴인식 분야에서 널리 사용되고 있을 뿐만 아니라 지문분류에서도 높은 성능을 보이고 있다. SVM은 이진클래스 분류기이기 때문에 다중클래스 문제인 지문분류를 위해서 적절한 분류기 생성과 결합 기법이 필요하며, 본 논문에서는 일대다(one-vs-all; OVA) 방식으로 구성된 SVM을 naive Bayes(NB) 분류기를 이용하여 동적으로 구성하는 분류방법을 제안한다. 지문분류에서 대표적으로 사용되는 특징인 FingerCode와 지문의 구조적 특징인 특이점과 의사융선을 사용하여 OVA SVM과 NB 분류기를 학습하고, 포섭구조의 분류기를 구성하여 효과적인 지문분류를 수행한다. NIST-4 데이타베이스에 제안하는 방법을 적용하여 5클래스 분류에 대해서 90.8%의 높은 분류율을 획득하였으며, OVA 전략의 SVM을 다중클래스 분류문제에 적용할 때 발생하는 동점문제를 효과적으로 처리하였다. Fingerprint classification reduces the number of matches required in automated fingerprint identification systems by categorizing fingerprints into a predefined class. Support vector machines (SVMs), widely used in pattern classification, have produced a high accuracy rate when performing fingerprint classification. In order to effectively apply SVMs to multi-class fingerprint classification systems, we propose a novel method in which SVMs are generated with the one-vs-all (OVA) scheme and dynamically ordered with na?ve Bayes classifiers. More specifically, it uses representative fingerprint features such as the FingerCode, singularities and pseudo ridges to train the OVA SVMs and na?ve Bayes classifiers. The proposed method has been validated on the NIST-4 database and produced a classification accuracy of 90.8% for 5-class classification. Especially, it has effectively managed tie problems usually occurred in applying OVA SVMs to multi-class classification.

Data Mining Techniques : More Accurate Classified Algorithm For Cardiopulmonary Diseases Prediction

Taher M. Ghazal,Syed Hakim Masood,Atif Ali,Muhammad Usama Nazir 한국차세대컴퓨팅학회 2022 한국차세대컴퓨팅학회 학술대회 Vol.2022 No.10

Data mining techniques develop a more accurate classification algorithm for patients classified as either normotensive, prehypertensive, or hypertensive. Logistic Model Tree, NBTree, and Bagging were chosen as the three classification models with tenfold cross-validation (LMT). Over 24 hours, we collected ABP readings from 1161 patients. To analyze the data, data mining techniques were used and a tool called WEKA. The data was analyzed based on age, gender, wake-up blood pressure, medication, sleep-up blood pressure, and overall blood pressure. According to bagging results, 886 cases (76.3 percent) are correctly classified, with 270 cases classified as pre-hypertensive, 436 cases as Normotensive, and 180 cases as hypertensive. NBTree's results show that 882 (75.9%) of the 1161 instances are correctly classified. Pre-hypertensive patients make up 256, normotensive patients 442, and hypertensive patients 184. Of the 1161 instances, the LMT algorithm correctly classified 878 (75.6 percent). According to the results, 275 people are pre-hypertensive, 431 are normotensive, and 172 are hypertensive. According to our findings, bagging is the most accurate classifier for the 24 hour ABP Monitoring dataset we used. Bagging achieves less overfitting because it focuses on global accuracy. It stabilizes and improves the accuracy of unstable methods compared to single classifiers.

A novel nomogram of naïve Bayesian model for prevalence of cardiovascular disease

Kang, Eun Jin,Kim, Hyun Ji,Lee, Jea Young The Korean Statistical Society 2018 Communications for statistical applications and me Vol.25 No.3

Cardiovascular disease (CVD) is the leading cause of death worldwide and has a high mortality rate after onset; therefore, the CVD management requires the development of treatment plans and the prediction of prevalence rates. In our study, age, income, education level, marriage status, diabetes, and obesity were identified as risk factors for CVD. Using these 6 factors, we proposed a nomogram based on a $na{\ddot{i}}ve$ Bayesian classifier model for CVD. The attributes for each factor were assigned point values between -100 and 100 by Bayes' theorem, and the negative or positive attributes for CVD were represented to the values. Additionally, the prevalence rate can be calculated even in cases with some missing attribute values. A receiver operation characteristic (ROC) curve and calibration plot verified the nomogram. Consequently, when the attribute values for these risk factors are known, the prevalence rate for CVD can be predicted using the proposed nomogram based on a $na{\ddot{i}}ve$ Bayesian classifier model.