0000000000255570
AUTHOR
Florin Stoica
A General Frame for Building Optimal Multiple SVM Kernels
The aim of this paper is to define a general frame for building optimal multiple SVM kernels. Our scheme follows 5 steps: formal representation of the multiple kernels, structural representation, choice of genetic algorithm, SVM algorithm, and model evaluation. The computation of the optimal parameter values of SVM kernels is performed using an evolutionary method based on the SVM algorithm for evaluation of the quality of chromosomes. After the multiple kernel is found by the genetic algorithm we apply cross validation method for estimating the performance of our predictive model. We implemented and compared many hybrid methods derived from this scheme. Improved co-mutation operators are u…
Using the Breeder genetic algorithm to optimize a multiple regression analysis model used in prediction of the mesiodistal width of unerupted teeth
For the prediction of the unerupted canine and premolars mesiodistal size, have been proposed different variants of multiple linear regression equations (MLRE). These are based on the amount of the upper and lower permanent incisors with a tooth of the lateral support. Aim of present study was to develop a method for optimization of MLRE, using a genetic algorithm for determining a set of coefficients that minimizes the prediction error for the sum of permanent premolars and canines dimensions from a group of young people in an area Romania's central city represented by Sibiu. To test the proposed method, we used a multiple linear regression equation derived from the estimation method propo…
Integrated Tool for Assisted Predictive Analytics
Organizations use predictive analysis in CRM (customer relationship management) applications for marketing campaigns, sales, and customer services, in manufacturing to predict the location and rate of machine failures, in financial services to forecast financial market trends, predict the impact of new policies, laws and regulations on businesses and markets, etc. Predictive analytics is a business process which consists of collecting the data, developing accurate predictive model and making the analytics available to the business users through a data visualization application. The reliability of a business process can be increased by modeling the process and formally verifying its correctn…
Building a new CTL model checker using Web services
This Computation Tree Logic (CTL) is widely used to capture compositions of reactive systems. Model checking is particularly well-suited for the automated verification of finite-state systems, both for software and for hardware. A CTL model checker tool allows designers to automatically verify that systems satisfy specifications expressed in the language of CTL logic. In this paper we present a new CTL model checker implemented in client-server paradigm. CTL Designer, the client tool, allows an interactive construction of the CTL models as state-transition graphs. Java and C# APIs are provided for programmatic construction of large models. The server part of our tool embeds the core of the …
ATL model checking in the cloud
This paper gives an overview of our recent work on implementing a new interactive ATL model checker for verification of open systems. In verification based on model checking, we need to provide a model of the system and also write down the properties (ATL formulas) that we require the system to satisfy. Traditionally, the semantics of ATL is given in terms of concurrent game structures. In contrast to previous approaches, our tool permits an interactive design of the ATL models as state-transition graphs, and is based on client/server architecture. The server part, published as Web service in OpenShift cloud platform, embeds the core of the ATL model checker, and the client provides an intu…
Implementing an ATL model checker tool using relational algebra concepts
Alternating-Time Temporal Logic (ATL) is a branching-time temporal logic that naturally describes computations of open systems. An open system interacts with its environment and its behavior depends on the state of the system as well as the behavior of the environment. ATL model-checking is a well-established technique for verifying that a formal model representing such a system satisfies a given property. In this paper we describe a new interactive model checker environment based on algebraic approach. Our tool is implemented in client-server paradigm. The client part allows an interactive construction of ATL models represented by concurrent game structures as directed multi-graphs. The se…
Verification of JADE Agents Using ATL Model Checking
It is widely accepted that the key to successfully developing a system is to produce a thorough system specification and design. This task requires an appropriate formal method and a suitable tool to determine whether or not an implementation conforms to the specifications. In this paper we present an advanced technique to analyse, design and debug JADE software agents, using Alternating-time Temporal Logic (ATL) which is interpreted over concurrent game structures, considered as natural models for compositions of open systems. In development of the proposed solution, we will use our original ATL model checker. In contrast to previous approaches, our tool permits an interactive or programma…
Optimization of Complex SVM Kernels Using a Hybrid Algorithm Based on Wasp Behaviour
The aim of this paper is to present a new method for optimization of SVM multiple kernels The kernel substitution can be used to define many other types of learning machines distinct from SVMs We introduced a new hybrid method which uses in the first level an evolutionary algorithm based on wasp behaviour and on the co-mutation operator LR−Mijn and in the second level a SVM algorithm which computes the quality of chromosomes The most important details of our algorithms are presented The testing and validation proves that multiple kernels obtained using our genetic approach are improving the classification accuracy up to 94.12% for the “leukemia” data set.
Prediction of the mesiodistal size of unerupted canines and premolars for a group of Romanian children: a comparative study
Objectives The aim of the present study was to develop an optimization method of multiple linear regression equation (MLRE), using a genetic algorithm to determine a set of coefficients that minimize the prediction error for the sum of permanent premolars and canine dimensions in a group of young people from a central area of Romania represented by a city called Sibiu. Material and Methods To test the proposed method, we used a multiple linear regression equation derived from the estimation method proposed by Mojers, to which we adjusted regression coefficients using the Breeder genetic algorithm. A total of 92 children were selected with complete permanent teeth with no clinically visible …