ABOUT ME INTRODUCTION DATA GATHERING DATA CLEANING EXPLORING DATA CLUSTERING ARM and NETWORKING DECISION TREES NAIVE BAYES SVM CONCLUSIONS INFOGRAPHIC

CONCLUSIONS

Given the increasing trend of suicide attempts and suicide rates among Black youth, this project took a deeper dive into the issue using various data. The main objective of this project was to closely identify trends in geographic areas where attempts of self-harm are the largest among Black youth, identify risk factors that contribute to such problem, and assign what support has been made available for Black youth. This project attempted to evaluate the problem and provide next steps to providing solutions. Using various dataset and machine learnings tools the key take ways are as following:

Expletory data analysis of YRBSS revealed that feelings and behaviors patterns such as physical fighting, carrying a weapon, and drug and alcohol consumption were highly correlated with thoughts and plans of suicide. While this analysis could not identify interesting patterns by race, interesting patterns by gender were observed. Female respondents were more likely than their male counterparts to report feelings of sadness and plans to attempt suicide. These findings are consistent with recent CDC finding that between 2019 and 2020, suspected suicide attempt ED visits were 50.6% higher among girls aged 12–17 years old.This indicates the need for gender-specific suicide prevention outreach efforts to ensure the well-being of young girls

Figure 1:Plans of Sucide by Gender
Image

Analysis from America's Health risk dataset provided insightful trends by geography. Using the four census regions - west, midwest, south, and northeast - high rates of teen suicide are reported in the western states. States in the west census region are Alaska, Arizona, California, Colorado, Hawai'i, Idaho, Montana, New Mexico, Nevada, Oregon, Utah, and Wyoming. These states typically have a lower population of Black people than states in the southern census region who have a large share of Black population, and have a relatively lower rate of teen suicide. These findings indicate the need to conduct further research on environmental patterns among states on the western census region share and how that is related to the high rates of teen suicide. This also calls for intentional campaigns and programs to combat teenage suicide in these states.

Figure 2: Teen Suicide by Census Region
Image

The association rule mining analysis revealed behaviors that commonly occur with those who have had thoughts or plans of attempting suicide. Some of the most interesting findings include that serious consideration of suicide, alcohol consumption, and frequently getting in physical fights were strongly associated behaviors. On the contrary, those who don't report using hallucinogenic drugs and said "none" regarding thoughts and plans of suicide. This indicated a strong relationship between drugs and alcohol consumption and thoughts of suicide.

Black people face an increased rates of risk factors, including experiences of racism, higher rates of unemployment and financial and food insecurity, disparities in other aspects of health, and limited access to care, all of which result in an increased burden of mental illness in Black communities. Black youth continue to be less likely to receive mental health treatment for depression when needed, compared to White youth. However these diffrences were not identified in our analysis.This may be due to the lack or mismatch of available data, flows in the models, or time restriction.

With more time and more access to data, Data science and algorithm will be critical tools to better predict suicided patterns among all age demographics. The given findings from this analysis call for better public health and suicide awareness outreach targeting young girls, youth in the west, and teens engaged in risky behaviors involving drugs, alcohol, and physical fights.

Quote of the day: "If you torture data long enough, it will tell you whatever you want to hear."