Constructing a System for Analysis of Machine Learning Techniques for Early Detection of Thyroid
Sayyad Rasheeduddin1, Kurra Rajasekhar Rao2
1Sayyad Rasheeduddin, Department of Computer Science and Engineering, India.
2Dr. Kurra Rajasekhar Rao, Department of Computer Science and Engineering, India.
Manuscript received on 01 November 2019 | Revised Manuscript received on 13 November 2019 | Manuscript Published on 22 November 2019 | PP: 1978-1981 | Volume-8 Issue-6S3 September 2019 | Retrieval Number: F13850986S319/19©BEIESP | DOI: 10.35940/ijeat.F1385.0986S319
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Thyroid is an unending and complex infection caused by unedifying levels of TSH (Thyroid Simulation Hormone) or by thyroid organ problems themselves. Hashimoto’s thyroid is the most widely recognized cause of hypothyroidism. The body makes anticorps that pulverize the thyroid organ in an auto-safe condition. It offers machine learning algorithms in the system proposed to predict thyroid disease in disease-intensive societies effectively. This is a serious concern for public health even though it is massively increasing in many countries. This shows that the problem must be predicted as urgently as possible to overcome the shortcomings of previously existing clinical decision-making tools with low precision. This paper examines numerous machine learning strategies for osteoporosis prediction. The paper examines and assesses the use of the strategy of feature selection combined with classification techniques. WEKA’s classification techniques are used to measure an osteoporosis data set. The results are calculated by means of various test options, including 10-fold cross-validation, training sets and the percentage divided with and without the selection method. The results are compared with correctly classified instances, runtime, kappa and absolute mean values for experiments with and with-out feature selection techniques
Keywords: Classification, Data Mining, Machine Learning, Decision Tree.
Scope of the Article: Machine Learning