Title of the Paper: Evaluation of Data Mining Classification Methodologies for Single Label Learning on benchmark data sets Using R and WEKA
Authors: Madana Mohana R, Satyanarayana V
Abstract:
The objective of this paper is to introduce, explain and compare the performance of single-labeled supervised learning algorithms in R language on benchmark single labeled data sets. Data is rich with hidden information that can be used for intelligent decision making. Classification is a form of data analysis that extracts models describing important data classes, the traditional classification algorithms like decision tree, random forest, support vector machine, naive-bayes are used under inspection. We have considered four measures (sensitivity, specificity, accuracy, F-measure) of performance here, the observations of all data set accuracy lead to infer that Random Forest outperforms the other classification methods. For more justification of our result we have implemented the same algorithms with same data sets in weka tool also.
Keywords: Decision Tree, Random Forest, SVM, Naive Bayes, R, Weka
Title of the Paper: Data Mining Techniques for CPD of Diabetes
Authors: N. Jayanthi , B . Vijay Babu, N. Sambasiva Rao
Abstract:
Disease diagnosis is very important for saving patient life. There are some diseases which are chronic for example today’s major and serious health problem is diabetes. This is known as modern society disease. Even though huge medical data is available absences of disease diagnosis keep a expert to opine about the grade of disease with confidence. Medical professionals need a prediction method to diagnose diabetes. In this situation Data Mining techniques are very useful for classification, prognosis and diagnosis of a disease. Early detection of diabetes in a patients help them for prevention of the disease to some extent. This papers gives a summary of old and recent techniques used for classification, prognosis and diagnosis of diabetes.
Keywords: Diabetes, data mining, classification, prediction, diagnosis
Title of the Paper: Machine Learning and Data Mining Techniques for Sign Language Recognition and Retrieval System
Authors: R. Madana Mohana, Dr. A. Rama Mohan Reddy
Abstract:
This paper presents a technique to acknowledge 32 American Sign Language distinctive letters and numbers from image signs, independence of signer and environment of image capture. Input pictures are mapped to the YCbCr color space, binarized and resized to 70×70 px. Principal Component Analysis is then performed on these binary images exploitation their pixels as features. This technique recognized signer-dependent signs with an accuracy of 100% and signer-independent signs with an accuracy of 62.37%, that will increase to 78.49% if dissimilar signs only used.
Keywords: SLR Systems; Image Processing; Principal Component Analysis, Support Vector Machine
Title of the Paper: Snowfall Prediction Techniques – A State of the Art Review
Authors: C Kishor Kumar Reddy, B Vijaya Babu
Abstract:
Snow carries the imperative role in the matter of aquatic, animal and human life in complete etiquette of climate circumstances and also considered to be one of the most natural climatic wonders whose nowcasting is arduous and challenging. In the mountainous areas, snow has a resilient impact on traffic, sewer systems. On the other hand, the snow hydrological cycle is one of the most amalgamated and challenging elements to distinguish and to model. This survey paper provides a diverse assortment of approaches which are epitomized by various academicians, researchers, scientists and meteorological departments in nowcasting snow. Hence the upshot of this paper is to show the available techniques and therefore comparison table exposes the accuracy and future scope of an individual methods used for nowcasting snowfall.
Keywords: Artificial Intelligence, Classification, Hydrology, Neural Networks, Nowcasting
Title of the Paper: Intelligent Street Light Monitoring And Control Using Micro Controller
Authors: R. Sathish Kumar
Abstract:
A novel procedure is actualized for the street light monitoring based autonomous and practical framework. The principle point of this venture is to reduce power consumption. The goal is to plan a device which characterizes a protocol to save more power compared to conventional systems. Here we are using ARM7 controller, ARM7 utilizes less power compared to existing controller. The proposed system uses the IOT based wireless devices which allow more efficient lamps management. The designed system uses sensors to control and guarantee the optimal system parameters The proposed system saves around 80.8% power for the outdoor street environment because of using sensors, LED lamps, IOT based communication technology.
Keywords: Embedded systems, IR sensors, LED lamps, IOT technology