Harikrishna S.,Vivekananda Institute of Technology and Science |
Farquad M.A.H.,University of Hyderabad
Advanced Materials Research | Year: 2012
Credit Scoring is the use of statistical/intelligent models to transform relevant data into numerical measures that guide the management and decision makers to make decisions such as accept/reject, pricing, pay/no pay and collections. This study focuses on predicting whether a credit applicant can be categorized as good or bad from the supplied data. Many researchers have recently worked on an ensemble of classifiers for such problems. It is observed from the literature that feature selection reduces the complexity of the system and improves the accuracy as well. Efficiency of SVM for feature selection and as a classifier in tandem and its application to credit scoring is analyzed in this paper. During the first step, SVM-RFE (Recursive Feature Elimination) is employed for feature selection and during the second step various architectures of SVM viz., Standard SVM, PSO-SVM and EVO-SVM are employed for classification purpose. The effectiveness of various approaches tested are evaluated using UK credit data and German credit data. It is observed that feature selection using SVM-RFE not only simplifies the process of credit scoring but also improves the accuracy of the system. © (2012) Trans Tech Publications, Switzerland.
Theja K.,Malla Reddy Engineering College Autonomous |
Mechineni P.,Vivekananda Institute of Technology and Science
ICACCS 2013 - Proceedings of the 2013 International Conference on Advanced Computing and Communication Systems: Bringing to the Table, Futuristic Technologies from Around the Globe | Year: 2014
Mobile Ad-Hoc Network is widely used network. It has many applications like military applications, search and rescue operations, conferences, law enforcement, etc. Most of these applications use multicasting protocol in Mobile Ad-Hoc Network (MANET). Initially Conventional flat multicasting protocols were used. A team affinity model has been used in this protocol wherein the multicast tree consists of teams, rather than individual nodes, and proved scalable and efficient for large scale mobile ad hoc networks. But it has some drawbacks (i.e.) Conventional flat multicasting protocols does not scale well to network size and are not suitable for team multicasting in large scale mobile ad hoc networks. In order to overcome these key problems of scalability to large membership as well as network size and fault tolerance to link/node failures. We propose a new protocol namely Fault Tolerant Hierarchical Multicast Routing Protocol (denoted as FTHMRP) which aims to develop a hierarchical multicast routing protocol based on hypercube architecture for team multicast with fault tolerance. FTHMRP constructs a hierarchy of team leaders and provides fault tolerant multicasting for data sensitive large scale mobile ad hoc networks. Simulation study shows that, FTHMRP provides efficient and reliable multicast compared with the application of a flat multicast scheme (ODMRP) that does not exploit team coordinated motion. © 2013 IEEE.
Reddy K.R.N.,Kakatiya Institute of Technology and Science |
Rao D.K.N.,Vivekananda Institute of Technology and Science |
Rao K.G.K.,Kakatiya University |
Kar K.K.,Indian Institute of Technology Kanpur
Journal of Composite Materials | Year: 2012
Treated and untreated woven century fiber (CF) composites are investigated for evaluating their flexural, tensile, and impact strengths, Barcol hardness, glass transition temperature, thermal resistance, and water absorption properties. Scanning electron microscopy study revealed a brittle fracture for treated fiber composite, while significant fiber pull-out is observed for the untreated fiber composite. Higher tensile and hardness properties, higher glass transition and thermal decomposition temperatures, and low water absorption properties are obtained for alkaline-treated composites. Enhanced interface bonding due to improved adhesion has helped in the formation of covalent bonds at the fiber-matrix interface, resulting in superior tensile, thermal, and water absorption properties of the treated fiber composite. Flexural and impact properties are higher for untreated fiber composite. This is due to the weak fiber-matrix interface and a differential strain in fiber and matrix. The weak interface has provided an energy absorbing mechanism and enhanced the impact energy absorption capacity. The tensile strength of CF composite is very much comparable with that of sisal fiber composite. The tensile and flexural properties are higher for the woven CF composites than those of short CF composites. But the impact properties are superior for short fiber composites compared to the woven fiber composites. Increased number of interfaces for the short fiber composites has contributed for the cushioning effect and an enhanced energy absorbing mechanism. Water absorption properties have improved for the treated fiber composites due to formation of a strong interface. © The Author(s) 2012 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav.
Almas A.,Taif University |
Farquad M.A.H.,Vivekananda Institute of Technology and Science |
Avala N.S.R.,Canara Bank |
Sultana J.,University of Hyderabad
7th International Conference on Digital Information Management, ICDIM 2012 | Year: 2012
Computational intelligence techniques are proved to be outperforming compared to standard statistical techniques, specifically when dealing with large, unbalanced and high dimensional data. In this paper we present an enhancement approach for improving the performance of decision tree using Support Vector Machine (SVM) when dealing with unbalanced data. The proposed approach modifies the available training data according to the predictions of SVM and this modified training data is then used to train decision tree. As the dataset at hand i.e. COIL data is highly unbalanced with 94:6 class distribution ratios, we also employed various standard sampling techniques for extensive analysis. Based on sensitivity measure, it is observed that the proposed approach enhanced the efficiency of decision tree exceptionally well. Other intelligent methods can be tested in place of DT. © 2012 IEEE.
Rampure P.B.,KLE University |
Venkataramana Reddy C.,Vivekananda Institute of Technology and Science |
Vijayakumar Reddy K.,JNTUH College of Engineering
ARPN Journal of Engineering and Applied Sciences | Year: 2012
The financial growth of the country is measured by efficient use of natural resources especially fuel. Fossil fuels have played a dominant role in the rapid industrialization of the world and thereby increased and improved quality of life. However, due to the threat of supply crunch ever rising prices and the effect of green house gases caused by conventional fuels there is an urgent need to explore the possibility of using straight vegetable oils (unmodified) as alternative fuels to reduce the pollution and to increase the energy self-reliance of the country, especially in rural areas. The study aims to investigate experimentally the performance analysis of non-edible rice bran oil, a straight vegetable oil [SVO] and its blends with diesel as fuel and also to investigate the exhaust emission characteristics in the single cylinder, direct injection, four stroke diesel engines typically used in Indian agricultural sector. Experiments were conducted by using the blends ofdiesel with non-edible rice bran oil in various proportions from ten to one hundred percentages by volume and at varying engine loads. Studies have found that the use of blends of more viscous non-edible rice bran oil and diesel in various percentages result in the decrease in the NOX emission while the brake thermal efficiency reduces marginally. © 2006-2012 Asian Research Publishing Network (ARPN).
Venkanna B.K.,Basaveshwar Engineering College |
Reddy C.V.,Vivekananda Institute of Technology and Science
International Journal of Alternative Propulsion | Year: 2012
The present work examines the use of a non-edible vegetable oil namely honne oil, a new possible alternative fuel for diesel engine. A direct injection (DI) diesel engine typically used in agricultural sector was operated on neat diesel (ND) and neat honne oil (H100). Static injection timing (start of static injection timing) was changed to study the performance, emission and combustion characteristics. It was observed that advancing the injection timing with H100 from the rated static injection timing (23° bTDC) increased the brake thermal efficiency and reduced CO, HC emissions and smoke opacity; however, NO x emission was increased. The ignition delay with H100 was higher than that with ND for all static injection timing under consideration. Improved premixed heat release rate was observed with H100 when the static injection timing is advanced however, for 28° bTDC premixed heat release rates decreased. The best static injection timing was found to be 27° bTDC for H100 based on brake thermal efficiency. Copyright © 2012 Inderscience Enterprises Ltd.
Venkanna B.K.,Basaveshwar Engineering College |
Venkataramana Reddy C.,Vivekananda Institute of Technology and Science
International Journal of Energy Technology and Policy | Year: 2013
The present work examines the use of a non-edible vegetable oil namely honne oil, a new possible source of alternative fuel for diesel engine. The use of honne oil (unheated and preheated) in an agricultural direct injection (DI) diesel engine has not been reported in the literature. The high viscosity of honne oil can be reduced by preheating (preheated to 60°C). The effect of temperature on the viscosity of honne oil has been investigated. Using preheated honne oil, brake thermal efficiency (BTE) and exhaust gas temperature (EGT) is increased compared to unheated honne oil. The emissions such as smoke opacity (SO), CO and HC of preheated honne oil were reduced and NOx emissions of preheated honne oil were increased compared to unheated honne oil. Premixed combustion phase of preheated honne oil is better than unheated honne oil and ND. Hence, preheated honne oil could be used in diesel engine for short-term application without any engine modification and adverse effect. Copyright © 2013 Inderscience Enterprises Ltd.
Venkanna B.K.,Basaveshwar Engineering College |
Venkataraman Reddy C.,Vivekananda Institute of Technology and Science
International Journal of Green Energy | Year: 2015
The present work examines the use of a non-edible oil namely honne oil, a new possible source of alternative fuel for diesel engines. The most detrimental parameter in the use of vegetable oil as diesel fuel is its higher viscosity. The viscosity of honne oil can be reduced by blending it with diesel fuel. Experiments have been conducted to evaluate the performance, emission, and combustion characteristics of a diesel engine typically used in agricultural sector over the entire load range when fuelled with honne oil/diesel fuel blends: H10 (10% honne oil + 90% diesel fuel) to H50. The engine performance, emission, and combustion characteristics of fuel blend H20 are found to be close to diesel fuel, whereas higher blend ratios are found to be inferior to diesel fuel. © 2015 Taylor & Francis Group, LLC.
Rentapalli V.R.,Vivekananda Institute of Technology and Science |
Khan Z.J.,Vivekananda Institute of Technology and Science
Communications in Computer and Information Science | Year: 2011
Evolution of wireless access technologies is about to the reach its fourth generation (4G). The adaptation of smart antenna techniques in future wireless systems is expected to have a significant impact on the efficient use of the spectrum and transparent operation across the multi technology wireless networks. With the rapid growth of wireless data traffic operators are anxious to quickly expand the capacity of their wireless networks. To address these 3GPP standards have incorporated powerful techniques for using so-called smart antennas. This paper focuses on the practical aspects of deploying smart antenna systems in the existing Radio Access Systems (RAS). Smart antenna techniques, such as multiple-input multiple-output (MIMO) systems, can extend the capabilities of 3G and 4G systems to provide customers with increased data throughput for mobile high-speed data applications. © 2011 Springer-Verlag.