Many people now a days are suffering from a thyroid disorder. Hence, the earlier detection of thyroid becomes very important in the medical field to cure the patient. Thyroid disorder was caused by the imbalanced state of thyroid hormones and is of two types: overproduction (Hyperthyroidism) and less production (Hypothyroidism). This paper applied machine learning models on the hypothyroid dataset taken from the UCI data repository to partially fulfil thyroid disorder detection. The detection methods are based on pattern recognition, machine learning and data mining. The experimental results have shown that the proposed model will give good results.