Fuzzy Systems and Knowledge Discovery: Third International Conference, FSKD 2006, Xi'an, China, September 24-28, 2006, Proceedings (Lecture Notes in Computer Science)

The basic text classification technique in forum application has been discussed in Sainin (2005a) and Sainin (2005b).The paper explains about the use of the basic naive Bayes algorithm to classify forum text me
ssages into two classes namely clean and bad.
In the problem of document text classification for conference paper, various papers themes will normally be classified manually by the conference management.Once the classification of the papers is ready, the parallel sessions for presentation according to the themes will be scheduled. This process is time consuming and may classify papers into unrelated themes.Based on this situation, an automated text document classification can replace the manual classification; hence reduce the decision time.In this paper, the similar algorithm that was applied in the previous experiment for the forum messages classification will be discussed according to the experiment for conference paper classification.

Text classification using Naive Bayes: An experiment to conference paper

Network troubleshooting is a significant process. Many studies were conducted about it. The first step in the troubleshooting procedures is represented in collecting information. It's collected in order to identify the problems. Syslog messages which are sent by almost all network devices include a massive amount of data that concern the network problems. Based on several studies, it was found that analyzing syslog data which can be a guideline for network problems and their causes. The detection of network problems can become more efficient if the detected problems have been classified based on the network layers. Classifying syslog data requires identifying the syslog messages that describe the network problems for each layer. It also requires taking into account the formats of syslog for vendors' devices. The present study aimed to propose a method for classifying the syslog messages which identify the network problem.This classification is conducted based on the network layers. This method uses data mining instrument to classify the syslog messages. The description part of the syslog message was used for carrying out the classification process.The relevant syslog messages were identified. The features were then selected to train the classifiers. Six classification algorithms were learned; LibSVM, SMO, KNN, Naive Bayes, J48, and Random Forest. A real data set was obtained from an educational network device. This dataset was used for the prediction stage. It was found that that LibSVM outperforms other classifiers in terms of the probability rate of the classified instances where it was in the range of 89.90%-32.80%. Furthermore, the validation results indicate that the probability rate of the correctly classified instances is >70%.

/pdf/comparative-analysis-of-classification-techniques-for-4z1l2cm2k7.pdf

Comparative analysis of classification techniques for network fault management

Data mining involves the use of data analysis tools to discover previously unknown, valid patterns and relationships in large data sets. As the use of internet is increasing day by day and with the advancement of internet news also publish online. So to handle this bulk amount of news various data mining techniques for classification had been used. In this paper we are using an intelligent system based on Hybrid algorithm (HMM, SVM and CART) for e-news classification. An intelligent system is designed which will extract the online news and then will find out category and subcategory wise news. System involves four main stages: a) Keyword Extraction b) Implementation of Hybrid Algorithm (HMM, SVM and CART). Data have been collected for experimentation from online newspapers like The Hindu, Hindustan Times and Times of India. The experimental results are based on the news categories and sub categories such as Entertainment: Bollywood 100% and Hollywood 90%, Sports: Cricket 90%, Football 90% and Hockey 78%, Matrimonial :Hindu 100% and Muslim 80%. In this paper we also compare the result of Hybrid algorithm (HMM, SVM and CART) with individual HMM and SVM Algorithm and conclude that Hybrid algorithm (HMM, SVM and CART) gave better result than that of what HMM and SVM individually gave.

Design and Implementation of Hybrid Algorithm for e-news Classification.

Network troubleshooting is an important process which has a wide research field. The first step in troubleshooting procedures is to collect information in order to diagnose the problems. Syslog messages which are sent by almost all network devices contain a massive amount of data related to the network problems. It is found that in many studies conducted previously, analyzing syslog data which can be a guideline for network problems and their causes was used. Detecting network problems could be more efficient if the detected problems have been classified in
terms of network layers. Classifying syslog data needs to identify the syslog messages that describe the network problems for each layer, taking into account the different formats of various syslog for vendors’ devices. This study provides a method to classify syslog messages that indicates the network problem in terms of network layers. The method used data mining tool to classify the syslog messages
while the description part of the syslog message was used for classification process. Related syslog messages were identified; features were then selected to train the classifiers. Six classification algorithms were learned; LibSVM, SMO, KNN, Naive Bayes, J48, and Random Forest. A real data set which was obtained from the
Universiti Utara Malaysia’s (UUM) network devices is used for the prediction stage. Results indicate that SVM shows the best performance during the training and prediction stages. This study contributes to the field of network troubleshooting, and the field of text data classification.

Network problems detection and classification by analyzing syslog data

Text classification (TC) is an important foundation of information retrieval and text
mining. The main task of a TC is to predict the text‟s class according to the type of tag given in advance. Most TC algorithms used terms in representing the document which does not consider the relations among the terms. These algorithms represent documents in a space where every word is assumed to be a dimension. As a result such representations generate high dimensionality which gives a negative effect on
the classification performance. The objectives of this thesis are to formulate algorithms for classifying text by creating suitable feature vector and reducing the dimension of data which will enhance the classification accuracy. This research combines the ontology and text representation for classification by developing five algorithms. The first and second algorithms namely Concept Feature Vector (CFV)
and Structure Feature Vector (SFV), create feature vector to represent the document.
The third algorithm is the Ontology Based Text Classification (OBTC) and is designed to reduce the dimensionality of training sets. The fourth and fifth algorithms, Concept Feature Vector_Text Classification (CFV_TC) and Structure Feature Vector_Text Classification (SFV_TC) classify the document to its related
set of classes. These proposed algorithms were tested on five different scientific paper datasets downloaded from different digital libraries and repositories. Experimental obtained from the proposed algorithm, CFV_TC and SFV_TC shown better average results in terms of precision, recall, f-measure and accuracy compared against SVM and RSS approaches. The work in this study contributes to exploring the related document in information retrieval and text mining research by using ontology in TC.

Suha Sahib Oleiwi

Papers

Enhanced ontology-based text classification algorithm for structurally organized documents