ملاذ رياض راضي البو العبره

With the increase in fatal accidents in Iraq, they have become a source of concern for both authorities and the public. Therefore, it has become necessary to conduct an analysis of these road accidents. This study aims to provide recommendations to responsible authorities after assessing the frequency of fatal traffic accidents and identifying the most common causes. This will provide actionable insights for decision-makers to formulate laws that allow for the reduction of these accidents and the reduction of human and economic losses. This paper applied data mining algorithms to three years of traffic fatal accident data in Iraq, excluding the Kurdistan Region. The results showed that people without driver's licenses and with primary school certificates were more likely to fail to wear seatbelts, making them a dangerous group. Married individuals aged 36-41 were also associated with fatal accidents. Based on the results, some recommendations were made to reduce these accidents.

Road traffic safety remains a pressing challenge globally, yet there is a scarcity of studies examining the specific factors contributing to traffic accidents in Iraq using advanced data mining techniques. This research aims to analyze traffic accident data in Iraq using data mining techniques to extract the factors most closely associated with accidents and provide practical recommendations for safe driving. The study utilizes the Iraq Traffic Accident Dataset (ITAD), which includes 10050 accident records but excludes data from the Kurdistan region, limiting the generalizability of findings to the entire country. The analysis focuses on variables such as accident type, weather conditions, visibility levels, and instances of driving under the influence. The Frequent Pattern Growth (FP-Growth) algorithm was employed to uncover association rules, selected for its computational efficiency in handling large, high-dimensional datasets and its ability to identify frequent patterns without candidate generation, unlike the Apriori algorithm. Classification trees were not used, as the study prioritizes exploratory association mining over predictive classification, aiming to reveal hidden relationships across multiple variables. The results highlight significant patterns: seatbelt usage is strongly associated with possessing a valid driving license, with individuals wearing seatbelts being more likely to hold valid licenses and be involved in accidents caused by moving vehicles. Furthermore, demographic factors reveal that married individuals using seatbelts and those uninjured in accidents are highly likely to have valid licenses. Conversely, individuals without licenses are often motorcycle owners, single, and possess only elementary school education, frequently neglecting seatbelt usage and contributing to crash-related accidents. Based on the discovered rules, this research proposes strategies to enhance traffic safety, such as tightening control over seatbelt use and increasing awareness about the importance of obtaining a driver's license, alongside targeted educational initiatives for at-risk groups. Despite the noted limitation in dataset coverage, these findings offer actionable insights for policymakers to reduce fatal traffic accidents in Iraq.