Digital Evangelism
  • Home
    • Big Data
    • Social Media
  • BLOG
  • RESOURCES
    • RESOURCE MENU >
      • ADVENTIST IDENTITY GUIDELINES
      • BIG DATA RESOURCES
      • BRANDING, IMAGE & DESIGN RESOURCES
      • CHURCH/MINISTRY SPECIFIC RESOURCES
      • COPYRIGHT & TRADEMARK BASICS
      • COURSES
      • EMAIL RESOURCES
      • GUIDANCE FOR HIRING SOCIAL MEDIA POSITIONS
      • PODCASTS
      • REPORTS & CASE STUDIES
      • SOCIAL MEDIA RESOURCES
      • (SOCIAL) VIDEO RESOURCES >
        • HOW TO START A VIDEO MINISTRY
      • TEXTING 4 CHURCHES
      • TRACKING & ANALTYICS
      • WATCH VIDEOS & TUTORIALS
      • WEBSITE TIPS
    • SOCIAL MEDIA GUIDELINES
  • SEO
    • SEO TERMS
  • Digital Discipleship & Evangelism
  • COVID-19 RESOURCES
  • eNEWSLETTER

Blog

#DigitalEvangelism

Discovering Mission-Oriented Patterns with Open Data in New York City

1/5/2017

4 Comments

 

Harvey Alférez, Ph.D 

Data Scientist, School of Engineering and Technology, Montemorelos University, Mexico
www.harveyalferez.com

Picture
Traffic in NYC as captured by my wife Doris’ camera lens
There is tons of open data on the Web. This data can be freely used by Seventh-day Adventists to try to figure out ways to help the inhabitants in the cities. This post describes how the students at my Pattern Recognition course at Montemorelos University and I have used open data and machine learning, which is a key component of data science, to discover interesting mission-oriented patterns for the church at NYC.

In my courses, I mostly focus on analyzing open data from NYC because of two reasons: 1) NYC has pivotal significance in our church’s ongoing Mission to the Cities project; and 2) NYC provides a portal that makes the wealth of public data generated by various NYC agencies and other city organizations available for public use [1].

Although the number of traffic deaths in NYC has fallen [2], city officials and traffic-safety groups agree that more aggressive steps must be taken to reach Mayor Bill de Blasio’s goal of eliminating traffic deaths in the city [3]. With this problem in mind, we analyzed a dataset of motor vehicle collisions in NYC, which is freely provided by the Police Department [4].

The studied dataset was created in 2014 and subsequently updated in 2016. This dataset registers motor vehicle collisions in Bronx, Brooklyn, Manhattan, Queens, and Staten Island from 2014 to 2016. This is a large dataset with 932,904 registered incidents! Moreover, each registered incident has 30 variables.

With traditional queries and spreadsheet analysis it is quite difficult (and sometimes impossible) to obtain timely answers to unseen patterns in large quantities of data, such as in our case study. In this kind of cases, machine learning, which “gives computers the ability to learn without being explicitly programmed” [5], can help us to grasp patterns we did not know that even exist.

From the set of 30 variables, we chose a subset to carry out the experiments. First, we chose the variables Date and Time because we wanted to know the day and time of each traffic incident. The Zip Code, Borough, Longitude, and Latitude variables were chosen because we wanted to know the demographic information of the accidents. Also, we had interest in figuring out the demographic groups that were injured the most. Therefore, we included in the experiments the Injured Persons, Injured Pedestrians, Injured Motorists and Injured Cyclists variables. Last but not least, we wanted to determine what provoked the accident and the type of vehicle that caused the accident. Therefore, we chose the Contributing Vehicle 1 and the Vehicle Type Code 1 variables from the dataset.

In order to analyze the data, we used Weka, which is a powerful tool for machine learning [6]. Although Weka contains a large range of machine learning algorithms, for our exploration we used the K-Means algorithm because the input data is unlabeled.  

Our findings are as follows:
  • On Fridays, around Prospect Park, Brooklyn, there are around 77,000 pedicab accidents registered in the dataset, which goes from 2014 to 2016. This finding can help Seventh-day Adventist congregations in the area (see the map below) to bring awareness to the community in order to prevent this kind of accidents. For example, pathfinders could go to the park on a field day and ride away with banners as flags on their bicycles or images and messages on shirts saying “Beware of your surroundings. Be swift on the brakes”. Also, information boots could be on display with pamphlets offering cycling safety tips to avoid collisions as well as what to do if people get involved in a traffic accident. Also a mini clinic to attend minor injuries in the vicinity would come in handy.
Picture
  • Alcohol has been one of the biggest accident contributors. Moreover, accidents in which bicycles were involved have caused the highest number of deaths and injuries. Church members could look for innovative ways to inform the general population of the dangers of consuming alcohol and driving, and about cycling safety.
  • On Thursdays, Fridays, and Saturdays, drivers tend to drive aggressively. This situation increases the number of accidents during those days. A solution the church could offer to this problem is to launch a social media campaign on stress management at the end of the week.
Let us use the knowledge that open data offers us to make a difference in the cities. As shown in the results above, big problems could have implementable simple solutions in which church members could make an extraordinary difference in their communities. Although a manual process could have been carried out to analyze the large dataset in our case study, it would have taken weeks or even months. In our case, the process just took a few days and a considerable low human-based analysis (computers did the hard work).

I thank the students at my Pattern Recognition course, Anthony, Claudia, Carlos, Isaías, Jairo, Eduard, Marco, Jaziel and Carlos, for their intense work on the experiments.
Picture

References:

1. The City of New York, “NYC Open Data,” (n.d.), https://data.cityofnewyork.us.

2. E. G. Fitzsimmons, “Number of Traffic Deaths in New York Falls for a Second Straight Year,” (2016), http://www.nytimes.com/2016/01/02/nyregion/number-of-traffic-deaths-in-new-york-falls-for-a-second-year-in-a-row.html.

3. M. Flegenheimer, “De Blasio Outlines Steps to Eliminate Traffic Deaths,” (2014), https://www.nytimes.com/2014/02/19/nyregion/de-blasio-unveils-plans-to-eliminate-traffic-deaths.html.

4. NYPD, “NYPD Motor Vehicle Collisions,” (2014), https://data.cityofnewyork.us/Public-Safety/NYPD-Motor-Vehicle-Collisions/h9gi-nx95. 

5. P. Simon, Too Big to Ignore: The Business Case for Big Data (Hoboken, NJ.: Wiley, 2013).

6. The University of Waikato, “Weka 3: Data Mining Software in Java,” (n.d.), http://www.cs.waikato.ac.nz/ml/weka/.
4 Comments
Angie
1/6/2017 08:53:40 pm

Well done! This is a great example of how the Church can respond better to the needs of their surrounding societies.
An infographic showing the results of some of those data analysis carried would give more insight to the reader of the complexity but power of your method.
Keep up the good work!

Reply
Jamie Schneider
1/7/2017 09:44:57 pm

Thanks! Good suggestion.

Reply
Harvey link
1/19/2017 08:49:00 am

Thank you for your suggestion, Angie! Let's keep in touch. God bless.

Reply
Paul Kujawa link
10/10/2022 02:34:13 pm

Also, we had interest in figuring out the demographic groups that were injured the most. Therefore, we included in the experiments the Injured Persons, Injured Pedestrians, Thank you for sharing your great post!

Reply



Leave a Reply.

    Archives

    August 2020
    July 2020
    June 2020
    April 2020
    March 2020
    February 2020
    December 2019
    November 2019
    August 2019
    July 2019
    June 2019
    May 2019
    April 2019
    March 2019
    February 2019
    January 2019
    December 2018
    November 2018
    October 2018
    September 2018
    July 2018
    June 2018
    May 2018
    April 2018
    March 2018
    February 2018
    January 2018
    December 2017
    November 2017
    October 2017
    September 2017
    July 2017
    June 2017
    May 2017
    April 2017
    March 2017
    February 2017
    January 2017
    December 2016
    November 2016
    October 2016

    Categories

    All
    A Allan Martin
    Adam Fenner
    Angeline Brauer
    Big Data
    Center For Online Evangelism
    Chris Matts
    Culture
    Data Science
    Data Tracking
    David Mwansa
    Design
    Digital Discipleship
    Digital Pew
    Digital Strategies Intern
    Dustin Comm
    Email Communications
    Erica Jones
    Event Registration
    Faith-hoyt
    Fundraising
    General
    Harvey Alférez
    Heidi Baumgartner
    James Gigante
    Jamie Schneider
    Jason Alexis
    Justin Khoe
    Kaleb Eisele
    Kimberly Luste Maran
    Marketing
    Member Care
    Michelle Diedrich
    Nick Wolfer
    Online Church
    Paul Hopkins
    Philanthropy
    Rachel Lemons Aitken
    Reaching Young Adults
    Rodney Brady
    SEO
    Social Media
    Stewardship
    Text Evangelism
    Video
    Volunteering
    Website
    Working From Home

    RSS Feed

Location

Hiding within those mounds of data is knowledge that could change the life of a patient, or change the world.

Atul Butte, Researcher
University of California, San Francisco

Contact Us

    Subscribe Today!

    We will use this email to send you updates on social media and big data initiatives.
Submit
Legal Notice   l   Privacy Policy
  • Home
    • Big Data
    • Social Media
  • BLOG
  • RESOURCES
    • RESOURCE MENU >
      • ADVENTIST IDENTITY GUIDELINES
      • BIG DATA RESOURCES
      • BRANDING, IMAGE & DESIGN RESOURCES
      • CHURCH/MINISTRY SPECIFIC RESOURCES
      • COPYRIGHT & TRADEMARK BASICS
      • COURSES
      • EMAIL RESOURCES
      • GUIDANCE FOR HIRING SOCIAL MEDIA POSITIONS
      • PODCASTS
      • REPORTS & CASE STUDIES
      • SOCIAL MEDIA RESOURCES
      • (SOCIAL) VIDEO RESOURCES >
        • HOW TO START A VIDEO MINISTRY
      • TEXTING 4 CHURCHES
      • TRACKING & ANALTYICS
      • WATCH VIDEOS & TUTORIALS
      • WEBSITE TIPS
    • SOCIAL MEDIA GUIDELINES
  • SEO
    • SEO TERMS
  • Digital Discipleship & Evangelism
  • COVID-19 RESOURCES
  • eNEWSLETTER