Exploring Common Symptoms in Patients with Respiratory Allergies Using K-Means Algorithm in the North-East of Iran in 2012-2015.

Background: As a common disease among people of almost any age, allergic rhinitis has many adverse effects such as lowering the quality of life and efficiency at work or school. Considering these conditions and the collection of large amounts of data, the present research was conducted on allergic rhinitis and asthma patients' data to extract the common symptoms of these diseases using cluster analysis and the k-means algorithm.

Materials and methods: The present cross-sectional research was conducted in Mashhad city. The inclusion criteria were affliction with one or two respiratory allergy diseases diagnosed by an allergy specialist through clinical history taking and physical examination. A researcher-made checklist was used in the present study for data collection. Then, the K-means algorithm's cluster analysis model was conducted to extract clusters (WEKA software (3, 6, 9)).

Results: Overall, 1,231 patients met the inclusion criteria. The result of the Cluster analysis consisted of Cluster 1 in allergic rhinitis consisted of 702 patients, and cluster 2 consisted of 382 patients.46 asthma patients were assigned to cluster 1 and 23 to cluster 2.Also, 60 asthma and allergic rhinitis patients were assigned to cluster 1 and 19 to cluster 2. The most common symptoms in all patients were rhinorrhea, sneezing, nasal congestion, and itchy nose.

Conclusion: Overall, Salsola kali was the most common allergen in allergic rhinitis and asthma patients. Also, the most common symptoms in patients are rhinorrhea, sneezing, itchy nose, and nasal congestion. This study can help physicians diagnose allergic rhinitis and asthma in geographical areas with a high prevalence of Salsola kali.