NYC Data Science Academy| Blog
Bootcamps
Lifetime Job Support Available Financing Available
Bootcamps
Data Science with Machine Learning Flagship ๐Ÿ† Data Analytics Bootcamp Artificial Intelligence Bootcamp New Release ๐ŸŽ‰
Free Lesson
Intro to Data Science New Release ๐ŸŽ‰
Find Inspiration
Find Alumni with Similar Background
Job Outlook
Occupational Outlook Graduate Outcomes Must See ๐Ÿ”ฅ
Alumni
Success Stories Testimonials Alumni Directory Alumni Exclusive Study Program
Courses
View Bundled Courses
Financing Available
Bootcamp Prep Popular ๐Ÿ”ฅ Data Science Mastery Data Science Launchpad with Python View AI Courses Generative AI for Everyone New ๐ŸŽ‰ Generative AI for Finance New ๐ŸŽ‰ Generative AI for Marketing New ๐ŸŽ‰
Bundle Up
Learn More and Save More
Combination of data science courses.
View Data Science Courses
Beginner
Introductory Python
Intermediate
Data Science Python: Data Analysis and Visualization Popular ๐Ÿ”ฅ Data Science R: Data Analysis and Visualization
Advanced
Data Science Python: Machine Learning Popular ๐Ÿ”ฅ Data Science R: Machine Learning Designing and Implementing Production MLOps New ๐ŸŽ‰ Natural Language Processing for Production (NLP) New ๐ŸŽ‰
Find Inspiration
Get Course Recommendation Must Try ๐Ÿ’Ž An Ultimate Guide to Become a Data Scientist
For Companies
For Companies
Corporate Offerings Hiring Partners Candidate Portfolio Hire Our Graduates
Students Work
Students Work
All Posts Capstone Data Visualization Machine Learning Python Projects R Projects
Tutorials
About
About
About Us Accreditation Contact Us Join Us FAQ Webinars Subscription An Ultimate Guide to
Become a Data Scientist
    Login
NYC Data Science Acedemy
Bootcamps
Courses
Students Work
About
Bootcamps
Bootcamps
Data Science with Machine Learning Flagship
Data Analytics Bootcamp
Artificial Intelligence Bootcamp New Release ๐ŸŽ‰
Free Lessons
Intro to Data Science New Release ๐ŸŽ‰
Find Inspiration
Find Alumni with Similar Background
Job Outlook
Occupational Outlook
Graduate Outcomes Must See ๐Ÿ”ฅ
Alumni
Success Stories
Testimonials
Alumni Directory
Alumni Exclusive Study Program
Courses
Bundles
financing available
View All Bundles
Bootcamp Prep
Data Science Mastery
Data Science Launchpad with Python NEW!
View AI Courses
Generative AI for Everyone
Generative AI for Finance
Generative AI for Marketing
View Data Science Courses
View All Professional Development Courses
Beginner
Introductory Python
Intermediate
Python: Data Analysis and Visualization
R: Data Analysis and Visualization
Advanced
Python: Machine Learning
R: Machine Learning
Designing and Implementing Production MLOps
Natural Language Processing for Production (NLP)
For Companies
Corporate Offerings
Hiring Partners
Candidate Portfolio
Hire Our Graduates
Students Work
All Posts
Capstone
Data Visualization
Machine Learning
Python Projects
R Projects
About
Accreditation
About Us
Contact Us
Join Us
FAQ
Webinars
Subscription
An Ultimate Guide to Become a Data Scientist
Tutorials
Data Analytics
  • Learn Pandas
  • Learn NumPy
  • Learn SciPy
  • Learn Matplotlib
Machine Learning
  • Boosting
  • Random Forest
  • Linear Regression
  • Decision Tree
  • PCA
Interview by Companies
  • JPMC
  • Google
  • Facebook
Artificial Intelligence
  • Learn Generative AI
  • Learn ChatGPT-3.5
  • Learn ChatGPT-4
  • Learn Google Bard
Coding
  • Learn Python
  • Learn SQL
  • Learn MySQL
  • Learn NoSQL
  • Learn PySpark
  • Learn PyTorch
Interview Questions
  • Python Hard
  • R Easy
  • R Hard
  • SQL Easy
  • SQL Hard
  • Python Easy
Data Science Blog > Python > New York City Weather and Vehicle Collision Data Analysis

New York City Weather and Vehicle Collision Data Analysis

Hua Yang
Posted on Feb 17, 2018

This work is a continuation of the previous work of New York City motor vehicle collision data visualization. While the previous work is focused on analysis of the collision data set only, in this work, I further include the weather data of New York City (NYC) and investigate their correlations.  Specifically, the objective of the project is to practice web scraping and data analysis using Python. I scraped 2013-2017 NYC hourly weather data from the โ€œWeather Undergroundโ€ website (high quality, reliable data by Central Park weather station), and studied correlations between the weather data and motor vehicle collisions data to find some good features to predict/model vehicle collisions.

Potential Business Values and Big Technical Plan

The ultimate outcome of the project would be a model to predict the number of vehicle collisions for the whole city, a borough or various different areas/spots of the city, etc., given weather and some other data sets, e.g. traffic volume data. The possible practical applications are:

  • โ€œSafe Routingโ€: if well built/performed, the collision model can be used to provide Safe Routing product/service for drivers on the road, and help yield less stressful driving experience.
  • โ€œProactive police arrangementโ€: the model can also be used for police department to proactively arrange police forces more on higher risk spots/areas than the others, to help better reduce/prevent collisions/risks.  

To achieve these business goals, a big technical project plan would include or further consider:

  • Besides weather, we need also consider road traffic (e.g. volume/intensity, history and real-time, etc.) information for effective collision modeling.
  • For better โ€œSafe Routingโ€ product/service, besides vehicle collisions, we may further consider NYC crime distribution data, etc.
  • All the factors of vehicle collisions and crime distribution, etc. need to be well trade-offed w/ total driving time/distance estimates for final routing decisions.

Web Scraping and Weather Data Set

I found and scraped high quality hourly NYC weather data on the โ€œWeather Undergroundโ€ website, followed its โ€œrobots.txtโ€ rule. The website is well organized and I mostly just followed lab-demoed scraping techniques using Scrapy, and didn't encounter much new/unexpected technical challenges in practice.

The finally collected data set is high quality and reliable NYC hourly weather data from 2013 to 2017  observed by Central Park weather station. The data set includes: 

  • 2 categorical variables:
    • Weather conditions: 16 detailed condition values are categorized/aggregated into: "clear", "cloudy", "fog", "rain", "snow", for analysis convenience and potentially better correlation result.
    • Wind direction: 18 detailed values are aggregated into "Calm", "Variable", "East", "South", "West", "North".
    • These two categorical variables are finally converted into dummy variables to be able to perform correlation analysis.
  • 6 numeric variables:
    • Visibility, temperature, humidity, pressure, dew point, wind speed (denoted respectively as "visi", "temp", "humi", etc. in the following result graphs). 

Data Preparation

Weather Data

To get the originally downloaded weather data set ready for analysis, I first did the following data preparation tasks.

  • Data cleaning: including: replacing undesirable strings (e.g. poorly typed/composed), dropping unwanted columns, renaming columns, changing column data types (from default Python Pandas read-in data type of "object" into appropriate types), sorting by index, etc.
  • Missing data handling: analyzed and properly handled missing data.
    • Dropped 4 columns with too many missing data (> 50%), e.g. "precipitation", "gust speed", "event", etc.
    • Imputed 6 numeric variables w/ time-interval linear interpolation from Pandas interpolate().
    • Imputed 2 categorical variables w/ simple forward fill and then backward fill
    • Studied more advanced imputing methods such as KNN and MICE, from "fancyimpute" package, but finally decided not using them, as they are very time consuming, and not necessary in the current situation.
  • Problem: some hours have more than one record, while some others have none. To solve this problem, I 
    • Grouped data into date-hours: picking mean and most common one for numeric and categorical variables, respectively.
    • Grouped motor vehicle collision data into hours and merged into weather data, imputed the missing hours weather data similarly as before.   

Vehicle Collision Data

To get collision data set ready for analysis, I did some similar work on data cleaning, and some different tasks as follows.

  • Problem: In the data set, there are 5 columns on causes, and 5 columns on involved types of vehicles, which are not easy to operate. To solve this issue:
    • Consolidated the columns into one single column for causes and vehicles, respectively. 
    • Used Python's Count() and most_common() function calls to pick the top 30 causes and top 15 vehicles (i.e. covering the majority of the total data) to re-define the desired categorical levels.
    • Converted top causes and top vehicles into dummy variables for correlation analysis.
  • Problem: there are quite a few columns on the number of injured, killed persons in terms of total, pedestrian, cyclist, and motorist, respectively. Similarly,  
    • Consolidated data: โ€œInjuredโ€ and โ€œKilledโ€ into one single โ€œVictimโ€ variable
    • Consolidated information: defined a new quantity: severity index = 1, if no hurt (i.e. num_injured and num_killed both are 0s), 2.55*num_injured (if num_injured not 0, and num_killed is 0), 7.55*num_of_killed (if num_killed not 0). To see if this new quantity has better correlation with other factors, than that of the simple original quantity of the total number of collisions.
  • Merge: grouped collision records into hours, and merged into hourly weather data frame.

Python Data Analysis

With the two data sets ready, I then calculated and drew collisions-per-hour frequency graphs to check different weather factorsโ€™ impact on collisions (using the basic Matplotlib plot).

To check correlations, I tried/tested w/ several different plotting tools.

  • Seaborn heatmap() vs Matplotlib matshow(): Prefer Seabornโ€™s heatmap for better looking overall. Moreover, it can put table/matrix numbers/values on top of the heatmap, which is more informative and sometimes very helpful.
  • Pandas scatter_matrix() vs Seaborn pairplot(): to check each individual variable density/histogram and each pair of variablesโ€™ scatter plot, prefer Pandas scatter_matrix(), for better rendering/looking.

For example, a nice looking Pandas scatter matrix plot is shown below, which shows scatter plots and kernel density estimates of the number of collisions and different weather factors between them and by each variable itself, respectively. We can see some clearly significant positive correlation exists between temperature and dew point.

Frequency Graphs

 

While "Clear" and "Overcast" have the two largest number of collisions, they are also occurred most of the time. Therefore, for a more fair check of each weather condition's impact on collisions, I divided total number of collisions by total number of occurring hours, i.e. the collisions per hour graph. Herein, heuristically, set minimum hours threshold = 10 (for now).

The more steep slopes/variations we see in a frequency graph, the higher impact that weather factor may have on collisions. Herein, roughly, we can see that: snow and humidity has more significant impact than that of temperature and visibility.

Note that herein in order to calculate total occurring hours of a continuous variable, we have to properly binning them so as to treat them the same way as for categorical variables. Also, the result of pressure and wind speed is not shown here, as the simple binning between min and max is not working well w/ them and I need better binning approach to handle many outliers of them. As for the result of dew point and wind direction, there is no much steep variation slope and hence no much potential impact observed, which is also omitted herein.  

Severity Index helps ?

The frequency graphs w/ the consolidated severity index quantity show a slight promising difference with those w/ the simple total number of collisions graphs, where we can see a little more steep variation herein than that w/ the number of collisions. However, whether or not or how much this much difference can help on modeling/prediction performance is a problem that definitely needs a careful check/comparison in actual modeling practice.   

Correlation Analysis

Overall, it is unfortunate that: from the graphs, we cannot see any significant correlation coefficients between different weather factors and collisions. Herein, the actual question is: how we should interpret this result properly. Esp, e.g. for snow factor, for which we observed significant impact in frequency graphs. 

In fact, this result of very low correlation coefficient is not so surprising, given the nature of the snow condition variable a dummy/binary variable, and occurring not very often in time. But then, how can we properly measure their relevant/concerned correlation impact on modeling? And how to effectively make use of these weather variables for better modeling?... All these questions/problems need further thinking/studies in the future.

To collect more insights either on collisions data itself or on its correlation with the weather factors, I also checked the correlation plots involving top collision causes and/or top involved vehicles. 

Some interesting observations are highlighted as follows.

  • High positive correlation exists between top causes/vehicles and collisions, and between top causes and top involved types of vehicles.
  • Some noticeable significant positive correlation spots are:
    • ("following too closely", "passing or lane usage improperly", "unsafe lane changing") among each other, and for them to be with ("reaction to other vehicle", "unsafe speed") and ("pedestrian/bicyclist error/confusion").
    • ("van", "other") and ("small commercial", "bus", "large commercial", "livery vehicle") with ("fatigued/drowsy", "lost consciousness", "prescription medication")
    • ("van", "other") with ("small commercial", "bus", "large comm")
  • Note: "taxi" is pretty much O.K. (i.e. not seen as problematic), with lower correlation with the other factors, while for "pick-up truck", we can see it's more problematic by having higher correlation w/ ("sport utility/station wagon", "oversized vehicle") and   ("following too closely", "passing or lane usage improperly", "unsafe lane changing").

As for the correlation plot between top involved types of vehicles and weather factors, some noticeable significant correlation spots are:

  • ("humidity") has negative correlation with top 8 causes ("Driver Inattention/Distraction", ..., "Prescription Medication")
  • ("temperature") has positive correlation with ("bicycle", "motorcycle")

What next?

In the future, the project can be further pursued in the following several directions.

  • Try further group hourly weather data into daily record, i.e. with higher value variations of variables, and re-check correlation to see if it gets any higher/more significant or not.
  • Further study how to effectively make use of binary dummy variables in modeling.
  • Try some modeling approaches and see the performance, anyway ...

Thank you! ๐Ÿ™‚ 

About Author

Hua Yang

Hua attended 12-week data science bootcamp of NYCDSA, which is really a great and awesome experience. Thanks a lot!
View all posts by Hua Yang >

Related Articles

Capstone
Catching Fraud in the Healthcare System
Capstone
The Convenience Factor: How Grocery Stores Impact Property Values
Capstone
Acquisition Due Dilligence Automation for Smaller Firms
Machine Learning
Pandemic Effects on the Ames Housing Market and Lifestyle
Machine Learning
The Ames Data Set: Sales Price Tackled With Diverse Models

Leave a Comment

Cancel reply

You must be logged in to post a comment.

No comments found.

View Posts by Categories

All Posts 2399 posts
AI 7 posts
AI Agent 2 posts
AI-based hotel recommendation 1 posts
AIForGood 1 posts
Alumni 60 posts
Animated Maps 1 posts
APIs 41 posts
Artificial Intelligence 2 posts
Artificial Intelligence 2 posts
AWS 13 posts
Banking 1 posts
Big Data 50 posts
Branch Analysis 1 posts
Capstone 206 posts
Career Education 7 posts
CLIP 1 posts
Community 72 posts
Congestion Zone 1 posts
Content Recommendation 1 posts
Cosine SImilarity 1 posts
Data Analysis 5 posts
Data Engineering 1 posts
Data Engineering 3 posts
Data Science 7 posts
Data Science News and Sharing 73 posts
Data Visualization 324 posts
Events 5 posts
Featured 37 posts
Function calling 1 posts
FutureTech 1 posts
Generative AI 5 posts
Hadoop 13 posts
Image Classification 1 posts
Innovation 2 posts
Kmeans Cluster 1 posts
LLM 6 posts
Machine Learning 364 posts
Marketing 1 posts
Meetup 144 posts
MLOPs 1 posts
Model Deployment 1 posts
Nagamas69 1 posts
NLP 1 posts
OpenAI 5 posts
OpenNYC Data 1 posts
pySpark 1 posts
Python 16 posts
Python 458 posts
Python data analysis 4 posts
Python Shiny 2 posts
R 404 posts
R Data Analysis 1 posts
R Shiny 560 posts
R Visualization 445 posts
RAG 1 posts
RoBERTa 1 posts
semantic rearch 2 posts
Spark 17 posts
SQL 1 posts
Streamlit 2 posts
Student Works 1687 posts
Tableau 12 posts
TensorFlow 3 posts
Traffic 1 posts
User Preference Modeling 1 posts
Vector database 2 posts
Web Scraping 483 posts
wukong138 1 posts

Our Recent Popular Posts

AI 4 AI: ChatGPT Unifies My Blog Posts
by Vinod Chugani
Dec 18, 2022
Meet Your Machine Learning Mentors: Kyle Gallatin
by Vivian Zhang
Nov 4, 2020
NICU Admissions and CCHD: Predicting Based on Data Analysis
by Paul Lee, Aron Berke, Bee Kim, Bettina Meier and Ira Villar
Jan 7, 2020

View Posts by Tags

#python #trainwithnycdsa 2019 2020 Revenue 3-points agriculture air quality airbnb airline alcohol Alex Baransky algorithm alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus ames dataset ames housing dataset apartment rent API Application artist aws bank loans beautiful soup Best Bootcamp Best Data Science 2019 Best Data Science Bootcamp Best Data Science Bootcamp 2020 Best Ranked Big Data Book Launch Book-Signing bootcamp Bootcamp Alumni Bootcamp Prep boston safety Bundles cake recipe California Cancer Research capstone car price Career Career Day ChatGPT citibike classic cars classpass clustering Coding Course Demo Course Report covid 19 credit credit card crime frequency crops D3.js data data analysis Data Analyst data analytics data for tripadvisor reviews data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization database Deep Learning Demo Day Discount disney dplyr drug data e-commerce economy employee employee burnout employer networking environment feature engineering Finance Financial Data Science fitness studio Flask flight delay football gbm Get Hired ggplot2 googleVis H20 Hadoop hallmark holiday movie happiness healthcare frauds higgs boson Hiring hiring partner events Hiring Partners hotels housing housing data housing predictions housing price hy-vee Income industry Industry Experts Injuries Instructor Blog Instructor Interview insurance italki Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter las vegas airport lasso regression Lead Data Scienctist Lead Data Scientist leaflet league linear regression Logistic Regression machine learning Maps market matplotlib Medical Research Meet the team meetup methal health miami beach movie music Napoli NBA netflix Networking neural network Neural networks New Courses NHL nlp NYC NYC Data Science nyc data science academy NYC Open Data nyc property NYCDSA NYCDSA Alumni Online Online Bootcamp Online Training Open Data painter pandas Part-time performance phoenix pollutants Portfolio Development precision measurement prediction Prework Programming public safety PwC python Python Data Analysis python machine learning python scrapy python web scraping python webscraping Python Workshop R R Data Analysis R language R Programming R Shiny r studio R Visualization R Workshop R-bloggers random forest Ranking recommendation recommendation system regression Remote remote data science bootcamp Scrapy scrapy visualization seaborn seafood type Selenium sentiment analysis sentiment classification Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau teachers team team performance TensorFlow Testimonial tf-idf Top Data Science Bootcamp Top manufacturing companies Transfers tweets twitter videos visualization wallstreet wallstreetbets web scraping Weekend Course What to expect whiskey whiskeyadvocate wildfire word cloud word2vec XGBoost yelp youtube trending ZORI

NYC Data Science Academy

NYC Data Science Academy teaches data science, trains companies and their employees to better profit from data, excels at big data project consulting, and connects trained Data Scientists to our industry.

NYC Data Science Academy is licensed by New York State Education Department.

Get detailed curriculum information about our
amazing bootcamp!

Please enter a valid email address
Sign up completed. Thank you!

Offerings

  • HOME
  • DATA SCIENCE BOOTCAMP
  • ONLINE DATA SCIENCE BOOTCAMP
  • Professional Development Courses
  • CORPORATE OFFERINGS
  • HIRING PARTNERS
  • About

  • About Us
  • Alumni
  • Blog
  • FAQ
  • Contact Us
  • Refund Policy
  • Join Us
  • SOCIAL MEDIA

    ยฉ 2025 NYC Data Science Academy
    All rights reserved. | Site Map
    Privacy Policy | Terms of Service
    Bootcamp Application