Analyzing Breath Alcohol Test Data in Ames, Iowa: Using Machine Learning

Ames, Iowa, which is the location of Iowa State University, encounters a substantial amount of alcohol-related occurrences, which exemplify the characteristics of a typical town centred around a college. This study analyses breath alcohol test data obtained by the Ames and Iowa State University Police Departments between January 2013 and December 2017. The dataset comprises data from 1,556 tests, offering a comprehensive perspective on the temporal trends of alcohol consumption patterns. We employ data analysis techniques to examine temporal patterns, demographic factors, and spatial dispersion of breath alcohol levels. The purpose of the findings is to provide information for public safety strategies and enhance the comprehension of alcohol consumption in college communities

Methodology

Data Collection and Preparation

  • The dataset “breath_alcohol_ames.csv” consists in breath alcohol test readings with columns covering year, month, day, hour, place, gender, and 2 test results (Res1 and Res2).
  • Loading Data: imported and handled the dataset for data manipulation and analysis using the Python pandas tool.

Cleaning Data

  • Examining the data rapidly allowed one to understand its structure and identify any missing or inconsistent
  • Managing Missing Values: Using appropriate estimation or elimination techniques will help to solve incomplete data problems so preserving the analytical
  • Data Consistency: Res1 and Res2 should ideally be precisely identical or very similar, thus their values should be coherent.

EDA (exploratory data analysis)

  • Examined test reading distribution over several years, months, days, and hours in order to identify trends over time.
  • Demographic Analysis: Investigated how gender affected breath alcohol levels in search of any discernible differences between male and female
  • Location Analysis: Examined test result spatial distribution between Iowa State University Police Department and Ames Police Department.

Statistical Interpretive Study

  • To provide an overview of the breath alcohol level data, computed descriptive statistics comprising the mean, median, and standard
  • Hypothesis Testing: To find relevant changes in breath alcohol levels depending on time, gender, and location, we ran statistical hypotheses

Data Collection and Preparation

  • Dataset: The dataset ‘breath_alcohol_ames.csv’ comprises breath alcohol test readings, encompassing columns for year, month, day, hour, location, gender, and two test results (Res1 and Res2).
  • Loading Data:Utilised the pandas library in Python to import and process the dataset for data manipulation and

Data Cleaning

  • Inspecting Data: Rapidly examined the data to comprehend its organisation and detect any absent or irregular values.
  • Handling Missing Values: Resolved the issue of incomplete data by utilising suitable imputation or elimination methods to maintain the accuracy of the
  • Data Consistency: Ensured coherence between the values of Res1 and Res2, as they should ideally be identical or very similar..

Exploratory Data Analysis (EDA)

  • Temporal Analysis: Examined the distribution of test readings across various time periods (years, months, days, and hours) to analyse trends over
  • Demographic Analysis: Examined the impact of gender on breath alcohol levels in order to detect any notable disparities between male and female
  • Location Analysis: Examined the spatial arrangement of test results between the Ames Police Department and Iowa State University Police

Statistical Analysis

  • Descriptive Statistics: Computed descriptive statistics, including the mean, median, and standard deviation, to provide a concise summary of the breath alcohol level data.
  • Hypothesis Testing: Performed statistical hypothesis tests to ascertain the presence of significant variations in breath alcohol levels with respect to time, gender, and

Visualization

  • Data Visualization: Employed the matplotlib and seaborn libraries to generate visualisations of the data, such as line charts, histograms, and bar
  • Trend Visualization: Analysed temporal patterns in breath alcohol levels to detect peak periods and fluctuations across different

Modeling and Prediction (if applicable)

  • Predictive Modeling: Created predictive models to anticipate future patterns in breath alcohol levels using past data (optional step depending on project scope).

Conclusion and Future Work

  • Results: Presented are the main discoveries, emphasising noteworthy patterns and demographic observations..
  • Implications: Explored the consequences of the findings for public safety strategies in Ames and other college
  • Future Research: Proposed avenues for additional investigation involve conducting a more detailed examination of particular occurrences and expanding the scope of the study to encompass other university .

Proposed avenues for additional investigation involve conducting a more detailed examination of particular occurrences and expanding the scope of the study to encompass other university communities

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Read More