Commenced in January 2007
Frequency: Monthly
Edition: International
Paper Count: 11

Search results for: Lucian N. Vintan

11 Meta-Classification using SVM Classifiers for Text Documents

Authors: Daniel I. Morariu, Lucian N. Vintan, Volker Tresp

Abstract:

Text categorization is the problem of classifying text documents into a set of predefined classes. In this paper, we investigated three approaches to build a meta-classifier in order to increase the classification accuracy. The basic idea is to learn a metaclassifier to optimally select the best component classifier for each data point. The experimental results show that combining classifiers can significantly improve the accuracy of classification and that our meta-classification strategy gives better results than each individual classifier. For 7083 Reuters text documents we obtained a classification accuracies up to 92.04%.

Keywords: Meta-classification, Learning with Kernels, Support Vector Machine, and Performance Evaluation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1378
10 Evolutionary Feature Selection for Text Documents using the SVM

Authors: Daniel I. Morariu, Lucian N. Vintan, Volker Tresp

Abstract:

Text categorization is the problem of classifying text documents into a set of predefined classes. After a preprocessing step, the documents are typically represented as large sparse vectors. When training classifiers on large collections of documents, both the time and memory restrictions can be quite prohibitive. This justifies the application of feature selection methods to reduce the dimensionality of the document-representation vector. In this paper, we present three feature selection methods: Information Gain, Support Vector Machine feature selection called (SVM_FS) and Genetic Algorithm with SVM (called GA_SVM). We show that the best results were obtained with GA_SVM method for a relatively small dimension of the feature vector.

Keywords: Feature Selection, Learning with Kernels, Support Vector Machine, Genetic Algorithm, and Classification.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1441
9 Feature Selection Methods for an Improved SVM Classifier

Authors: Daniel Morariu, Lucian N. Vintan, Volker Tresp

Abstract:

Text categorization is the problem of classifying text documents into a set of predefined classes. After a preprocessing step, the documents are typically represented as large sparse vectors. When training classifiers on large collections of documents, both the time and memory restrictions can be quite prohibitive. This justifies the application of feature selection methods to reduce the dimensionality of the document-representation vector. In this paper, three feature selection methods are evaluated: Random Selection, Information Gain (IG) and Support Vector Machine feature selection (called SVM_FS). We show that the best results were obtained with SVM_FS method for a relatively small dimension of the feature vector. Also we present a novel method to better correlate SVM kernel-s parameters (Polynomial or Gaussian kernel).

Keywords: Feature Selection, Learning with Kernels, SupportVector Machine, and Classification.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1471
8 Evaluating some Feature Selection Methods for an Improved SVM Classifier

Authors: Daniel Morariu, Lucian N. Vintan, Volker Tresp

Abstract:

Text categorization is the problem of classifying text documents into a set of predefined classes. After a preprocessing step the documents are typically represented as large sparse vectors. When training classifiers on large collections of documents, both the time and memory restrictions can be quite prohibitive. This justifies the application of features selection methods to reduce the dimensionality of the document-representation vector. Four feature selection methods are evaluated: Random Selection, Information Gain (IG), Support Vector Machine (called SVM_FS) and Genetic Algorithm with SVM (GA_FS). We showed that the best results were obtained with SVM_FS and GA_FS methods for a relatively small dimension of the features vector comparative with the IG method that involves longer vectors, for quite similar classification accuracies. Also we present a novel method to better correlate SVM kernel-s parameters (Polynomial or Gaussian kernel).

Keywords: Features selection, learning with kernels, support vector machine, genetic algorithms and classification.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1310
7 Using Suffix Tree Document Representation in Hierarchical Agglomerative Clustering

Authors: Daniel I. Morariu, Radu G. Cretulescu, Lucian N. Vintan

Abstract:

In text categorization problem the most used method for documents representation is based on words frequency vectors called VSM (Vector Space Model). This representation is based only on words from documents and in this case loses any “word context" information found in the document. In this article we make a comparison between the classical method of document representation and a method called Suffix Tree Document Model (STDM) that is based on representing documents in the Suffix Tree format. For the STDM model we proposed a new approach for documents representation and a new formula for computing the similarity between two documents. Thus we propose to build the suffix tree only for any two documents at a time. This approach is faster, it has lower memory consumption and use entire document representation without using methods for disposing nodes. Also for this method is proposed a formula for computing the similarity between documents, which improves substantially the clustering quality. This representation method was validated using HAC - Hierarchical Agglomerative Clustering. In this context we experiment also the stemming influence in the document preprocessing step and highlight the difference between similarity or dissimilarity measures to find “closer" documents.

Keywords: Text Clustering, Suffix tree documentrepresentation, Hierarchical Agglomerative Clustering

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1681
6 Working Capital Efficiency and Firm Profitability – Nigeria and Kenya

Authors: Lucian J. Pitt

Abstract:

The primary purpose of this study is to understand the differences in the relationship between working capital management efficiency, working capital investment decisions and working capital finance decisions and the profitability of firms within the context of two African developing economies, Kenya and Nigeria. The study finds that there is a significant difference in the relationship between the firm’s profitability and the working capital variables which suggests different challenges for working capital management in each of these countries.

Keywords: Working Capital Management, Working Capital Investment, Working Capital Finance, Profitability, Cash Conversion Cycle.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2765
5 The Benefits of IFRS Adoption – A Survey of Chief Financial Officers of Romanian Listed Companies

Authors: Lucian Munteanu

Abstract:

The move towards internationalization of accounting encountered a great boost, when in 2002 EU delegated the IASB to provide the accounting standards to be applied inside its frontiers. Among the incentives of the standardization of accounting on the international level, is the reduction of the cost of capital. Romania made the move towards IFRS before EU, when the country was not yet a member of it. Even if this made Romania a special case, it was scarcely approached. The leak of real data is usually the reason for avoiding. The novelty of this paper is that it offers an insight from the reality of Romanian companies and their view regarding the IFRS. The paper is based on a survey that the authors made among the companies listed on the first two tiers of the Bucharest Stock Exchange (BSE), which are basically, the most important companies in the country.

Keywords: Cost of capital, IFRS, information asymmetry, transparency.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2484
4 Model Discovery and Validation for the Qsar Problem using Association Rule Mining

Authors: Luminita Dumitriu, Cristina Segal, Marian Craciun, Adina Cocu, Lucian P. Georgescu

Abstract:

There are several approaches in trying to solve the Quantitative 1Structure-Activity Relationship (QSAR) problem. These approaches are based either on statistical methods or on predictive data mining. Among the statistical methods, one should consider regression analysis, pattern recognition (such as cluster analysis, factor analysis and principal components analysis) or partial least squares. Predictive data mining techniques use either neural networks, or genetic programming, or neuro-fuzzy knowledge. These approaches have a low explanatory capability or non at all. This paper attempts to establish a new approach in solving QSAR problems using descriptive data mining. This way, the relationship between the chemical properties and the activity of a substance would be comprehensibly modeled.

Keywords: association rules, classification, data mining, Quantitative Structure - Activity Relationship.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1560
3 Novel Glycopolymers Containing Carbohydrate Moiety: Copolymerization and Thermal Properties

Authors: Liliana M. Ştefan, Ana M. Pană, Geza Bandur, Marcel Popa, Lucian M. Rusnac

Abstract:

Polymers are one of the most widely used materials in our every day life. The subject of renewable resources has attracted great attention in the last period of time. New polymeric materials derived from renewable resources, like carbohydrates draw attention to public eye especially because of their biocompatibility and biodegradability. The aim of our paper was to obtain environmentally compatible polymers from monosaccharides. Novel glycopolymers based on D-glucose have been obtained from copolymerization of a new monomer carrying carbohydrate moiety with methyl methacrylate (MMA) via free radical bulk polymerization. Differential scanning calorimetry (DSC) was performed in order to study the copolymerization process of the monomer into the chosen co-monomer; the activation energy of this process was evaluated using Ozawa method. The copolymers obtained were characterized using ATR-FTIR spectroscopy. The thermal stability of the obtained products was studied by thermogravimetry (TG).

Keywords: DSC, glycopolymer, monosaccarides, TG.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1444
2 Long Term Variability of Temperature in Armenia in the Context of Climate Change

Authors: Hrachuhi Galstyan, Lucian Sfîcă, Pavel Ichim

Abstract:

The purpose of this study is to analyze the temporal and spatial variability of thermal conditions in the Republic of Armenia. The paper describes annual fluctuations in air temperature. Research has been focused on case study region of Armenia and surrounding areas, where long–term measurements and observations of weather conditions have been performed within the National Meteorological Service of Armenia and its surrounding areas. The study contains yearly air temperature data recorded between 1961- 2012. Mann-Kendal test and the autocorrelation function were applied to detect the change trend of annual mean temperature, as well as other parametric and non-parametric tests searching to find the presence of some breaks in the long term evolution of temperature. The analysis of all records reveals a tendency mostly towards warmer years, with increased temperatures especially in valleys and inner basins. The maximum temperature increase is up to 1,5°C. Negative results have not been observed in Armenia. The patterns of temperature change have been observed since the 1990’s over much of the Armenian territory. The climate in Armenia was influenced by global change in the last 2 decades, as results from the methods employed within the study.

Keywords: Air temperature, long-term variability, trend, climate change.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1818
1 Automat Control of the Aircrafts- Lateral Movement using the Dynamic Inversion

Authors: Mihai Lungu, Romulus Lungu, Lucian Grigorie

Abstract:

The paper presents a new system for the automat control of the aircrafts- flight in lateral plane using the cinematic model and the dynamic inversion. Starting from the equations of the aircrafts- lateral movement, the authors use two axes systems and obtained a control law that cancels the lateral deviation of the flying objects from the runway line. This system makes the aircrafts- direction angle to follow the direction angle of the runway line. Simulations in Matlab/Simulink have been done for different aircraft-s initial points and direction angles. The inconvenience of this system is the long duration of the “transient regime". That is why this system can be used independently, but the results are not very good; thus, it can be a part (subsystem) of other systems. The main system that cancels the lateral deviation from the runway line is based on dynamic inversion and uses, as subsystem, the control system for the lateral movement using the cinematic model. Using complex Matlab/Simulink models, the authors obtained the time evolution of the direction angle and the time evolution of the aircraft lateral deviation with respect to the runway line, for different values of the initial direction angle and for different wind types. The system has a very good behavior for all initial direction angles and wind types.

Keywords: Direction angle, Dynamic inversion, Lateraldeviation, Lateral movement

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1682