Diagnosis of ADHD using SVM algorithm

Anuradha, J.; Tisha,; Ramachandran, Varun; Arulalan, K. V.; Tripathy, B. K.

doi:10.1145/1754288.1754317

Cited by 23 publications

(12 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The accuracy of the training group is 100% while the accuracy of the testing group is 95.83%. The testing accuracy is much higher than the result of J. Anuradha et al [13] which is also classified by SVM [14]. And the comparison result is shown in Fig.4.…”

Section: B Classificationmentioning

confidence: 70%

RISC'13 Workshop: Keynote speaker 5

Liu

Gao

2013

2013 IEEE Third International Conference on Information Science and Technology (ICIST)

View full text Add to dashboard Cite

this paper analyzes the resting state fMRI signal of 21 ADHD subjects and 27 healthy volunteers, and proposes a novel method for extracting an effective feature in frequency domain. Utilizing this feature, the ADHD subjects and the control persons are classified with an accuracy of 95.83% by support vector machine (SVM). Furthermore, using this method, some specific brain regions such as the right amygdaloid nucleus, the left thalamus, cerebellum and vermis, with high classification accuracies, are relative to the pathological mechanism of ADHD which are consistent with the previous research results.

show abstract

Section: B Classificationmentioning

confidence: 70%

RISC'13 Workshop: Keynote speaker 5

Liu

Gao

2013

2013 IEEE Third International Conference on Information Science and Technology (ICIST)

View full text Add to dashboard Cite

show abstract

“…It is a kind of Pareto optimization problem having set of solutions and no other better alternative solution. They are characterized by trade-offs leading to multitude of Pareto optimal solutions [7]. The Pareto optimal solutions are determined based on the dominant and non-dominant solutions called Pareto front.…”

Section: Multi-objective Psomentioning

confidence: 99%

“…Recently, much work has been done on a multi-objective approach for FS in classification using PSO, referred to as MOPSO [22]. This approach uses two-level Pareto front FS algorithms [7,23] namely; non-dominated sorting PSO (NSPSO) and CMDPSO which uses the concepts of crowding, mutation and dominance.…”

Section: Pso In Feature Selectionmentioning

confidence: 99%

Uncertainty Based Hybrid Particle Swarm Optimization Techniques and Their Applications

Anuradha

Tripathy

2015

Multi-Objective Swarm Intelligence

Self Cite

View full text Add to dashboard Cite

In order to handle uncertainty in data, several uncertainty based models have been introduced. Perhaps the two most notable models among these are the fuzzy set introduced by Zadeh in 1965 and rough set introduced by Pawlak in 1982. These two models address two different aspects of uncertainty and these are complementary by nature. As a result their hybridization leads to better models. Particle Swarm Optimization (PSO) is an optimization technique that performs optimized search in the solution space for optimization by updation. In this chapter, we discuss on different types of PSOs and their application in classification, feature selection and rule generation. Further, we present several hybridization of PSO with fuzzy approach, rough approach and rough fuzzy approach in developing classification algorithms. Also, we discuss on a dynamic clustering algorithm which uses a rough fuzzy hybrid model embedded with PSO. We provide as an illustration the results of application of this algorithm on several real life data sets and provide its superiority through the computation of several index values for measuring classification accuracy like DB, D, α, ρ, α * and γ . Also, we compile some other applications of PSO.

show abstract

“…The most important advantage of applying the SVM algorithm is that it can control the complexity of the diagnostic process. This method was tested on children between the ages six to eleven years old and the results indicated a percentage of 88,674% success in diagnosing [14].…”

Section: Diagnosismentioning

confidence: 99%