Thyroid disorders occur due to dysfunction of the thyroid gland or pituitary gland, iodine deficiency, cancer in some parts of the body, or due to side-effects from other medications. Hyperthyroidism, Hypothyroidism, Goitre, and Thyroid cancer are some of the ailments that result due to thyroid disorders. Some other reasons like pregnancy, or medications for other illnesses may also show abnormal levels of thyroid hormones This research study aims to identify conditions based on which we could predict the type of thyroid disorder in patients. This could help in further diagnosis and treatment. We study various attributes commonly found in patients with thyroid disorders to identify those attributes that may specifically describe the type of thyroid disorder in a person. Moreover we analyze six different classes of thyroid disorders, their symptoms and try to classify what kind of disorder a person has based on the symptoms. Totally 1535 records with 29 attributes are taken for the study. Statistical techniques are used to analyze the frequency of occurrence of various factors towards each type of the disease and test for significance of factors is also done. The results are used to build a data model that helps to predict the occurrence of specific type of thyroid disorder in a patient based on significant symptoms. The results emphasize that age, sex, values of hormones like TSH, T3, T4 and FTI of a patient play a predominant role in classifying and determining the type of thyroid disorder in the person. We also classify the given dataset using various decision tree techniques in different ways and compare the results.

Data mining, Thyroid disorder, Classification, Prediction

[1]. Dr.G.Rasitha Banu, M.Baviya, “Predicting Thyroid Disease using Data Mining Technique”, International Journal of Modern Trends in Engineering and Research, e-ISSN 2349-9745, Pages 666-670, Vol 2(3), March 2015.
[2]. Irina Ionitak Liviu Ionita, “Prediction of Thyroid Disease Using Data Mining Techniques”, BRAIN. Broad Research in Artificial Intelligence and Neuroscience. Vol.7. pp.115-124. Nov 18, 2017
[3]. Prerana, Parveen Sehgal, Khushboo Taneha, “Predictive Data Mining for Diagnosis of Thyroid Disease using Neural Network”, International Journal of Research in Management, Science & Technology, e-ISSN 2321-3264, Vol 3(2), April 2015
[4]. Ebru Turanoglu Bekar, Gozde Ulutagay, Suzan Kantarci Savas, “Classification of Thyroid Disease by Using Data Mining Models: A comparison of Decision Tree algorithms.” The Oxford Journal of Intelligent Decision and Data Science, Vol 2016(2), Pages 13-28, doi: 10.5899/2016/ojids-00002,
[5]. K.Rajam, R. Jemina Priyadarsini, “A Survey on Diagnosis of Thyroid Disease using Data Mining Techniques”, International Journal of Computer Science and Mobile Computing, ISSN 2320-088x , Vol 5(5), pg 354-358, May 2016
[6]. Dr.B.Srinivasan, K.Pavya, “Diagnosis of Thyroid Disease Using Data Mining Techniques: A study”, International Research Journal of Engineering and Technology, e-ISSN 2395-0056, Vol 3(11), Nov 2016.
[7]. Dr. G.Rasitha Banu, M.Baviya, Dr. Murtaza Ali, “A Study on Thyroid Disease using Data Mining Algorithm”, International Journal of Technical Research and Applications, e-ISSN 2320-8163, Vol 3(4), PP 376-379, Aug 2015
[8]. Zoya Khalid, Sheema Sameen, Shaukat I Malike, Shehzad S, “Computational Analysis on the Role of GPM6A in human thyroid cancer”, Journal of Data Mining in Genome Proteinomics 3:114, doi: 10.4172/2153-0602.1000114, ISSN 2153-0602,Jan 2012.
[9]. B.Jothi, S.KRishnaveni, J. Jeyasudha, “Analysis of thyroid syndrome using K-Means Clustering Algorithm”, Journal of Chemical and Pharmaceutical Sciences, ISSN 0974-2115,
[10]. K. Saravana Kumar, Dr. R. Manicka Chezian, “Analysis on Suspicious Thyroid Recognition using Association Rule Mining”, Journal of Global research in Computer Science, ISSN 2229-371X, Vol. 3(9), Sep 2012.
[11]. Limin Wang, FangYuan Cao, ShuangCheng Wang, MingHui Sun, LiYan Dong, “Using k-dependence causal forest to mine the most significant dependency relationships among clinical variables for thyroid disease diagnosis”, PLOS One, Vol 12(8), e0182070, 2017, doi: 10.1371/journal.pone.0182070
[12]. Kevin M.Pantalone et al., “Measurement of Serum Free Thyroxine Index may provide additional case detection compared to free thyroxine in the diagnosis of central hypothyroidism”, Case Reports in Endocrinology, Vol 2015, Dec 8, 2015, doi: 10.1155/2015/965191