Can you call the EMPD?

So, I got a speeding ticket notification via my bank. When I investigated it I found out it was in Ekurhuleni. I checked the rest of the details of the ticket. On the date I had not driven in the Ekurhuleni Municipality. Following this I got access to the speeding fine pictures, lo and behold the car was a bakkie (which I do not own). Then I noticed that the number plate was almost identical to my car except for one letter. So the automatic number plate reader had made a mistake. Whew. Now how to fix this? Let me get hold of the EMPD traffic office. Easy right. Well, this has now turned into a nightmare that keeps on going.

TLDR: I still have not gotten hold of and EMPD office after 43 all,s 20 different unique numbers. Its getting a bit frustrating and perplexing. 

Here I present all the calls I have made to try to resolve this issue. My biggest problem is that I have not been to talk to anyone at EMPD. I have not at any time been able to connect to an EMPD office. Even after being given so many different numbers be 3 main call centers the City of Ekurhuleni runs.

Ekurhuleni Calls

Ekurhuleni Calls

Where do we start? At the beginning of course, asking how to get contact with EMPD on Twitter

Well. Got two numbers. Lets try both. Well .....

Well this was just the beginning. Finally I got a call back from one of the numbers, who then told me that done deal with traffic complaints (odd, the city Twitter account shared their number with a notice that it is for traffic complaints). Anyway, got a number that I would later learn is for the emergency services. I felt completly bad when I found this out, you dont want to misuse an emergency number. Well, story gets even more interesting. The Emergency center could not guarantee that any of the numbers they gave me would work. Meaning they knew there is a problem with numbers. Just FYI these are the only contacts the city shares on their website.

Ekurhuleni Contacts

Ekurhuleni Contacts

Did it get any better from here. No. Not by a long shot. I also got in contact with the official City of Ekurhuleni call center where the operator, matter of factly, told me they do not keep any numbers for EMPD. They only have municipal office numbers. They proceeded to give me a list of numbers. You now should know the pattern, none of the phone numbers worked. None, no joke.

Here I am a week later. I still have a fine for a car I don't own. I cant get in touch with EMPD. Help.

You can view my call data here

Posted in Ramblings Tagged with: , , , ,

NIPS Authors 2006 - 2016

As part of the Deep Learning Indaba, we have gotten a dataset of the accepted papers from NIPS 2006 - 2016 and countries of corresponding authors. We wanted to understand the scale of the non-participation of countries/regions as well as overall trends. The main post on this is now available on the Deep Learning Indaba page: Missing Continents: A Study using Accepted NIPS Papers. This post is to just add some other visualisations that might provide some more context. Its a short post.

Number of Accepted NIPS papers 2006 - 2016

Number of Accepted NIPS papers 2006 - 2016

Total Authors per Country [2006-2016]

Total Authors per Country [2006-2016]

Gratitude to all of those we spoke to during this exploration and also acknowledging the work of Emily Muller who we worked with on the visualisations.


The data was prepared by NIPS. There were some challenges with using it directly. NIPS did not use ISO country names. Some countries we had to check from the original papers. Authors seem to primarily be counted agains the country of institution, but we did see a few cases where the author country was not the institution country and NIPS used the author country. We still do believe that the analysis is interesting. 

Posted in CSIR, Data Science, Work Tagged with: , , ,

Data Science Africa 2017

Earlier in the 2017 I attended the UN Data Forum. Outside some oddities out of some of the things discussed at that conference (especially on Data and power dynamics, you can read a paper we wrote in response -> A critical and ecosystem consideration of data for sustainable development in Africa), I got the opportunity to meet John Quinn who is one of the organisers of Data Science Africa. I was pretty much intrigued about the summer school as well as workshops and there and then promised that I would find a way to make it to Tanzania in July for the 2017 version.

The good news, and big thanks, is that through a NRF Knowledge Interchange and Collaboration grant I was able to attend (specifically the workshops). Another thank you to CSIR for also being flexible and encouraging with this part of my work.

My experience presenting at the workshop

I got an opportunity to speak about the work our group does at the CSIR. Specifically our interesting journey with Social Media mining and Public Safety. I enjoyed the conversations leading up to the presentation because this was one of those conferences where one feels "these are my people". The talks prior had already touched on some of the themes I was going to talk about, so it made it easy to cover the work that the wider group is doing carrying on from our past and highlight the thrusts that our PhD, MS and Honours students are doing in the area. I also participated in a panel on social media data later in the day. So in the spirit of the show Review: 5/5 ?.

My experience learning from the workshops

It was such a breath of fresh air. Seriously. The topics were varied, the experiences shared and more discussion happened later in the day as one tried to get to the bottom of some of the areas discussed. I find it hard to point out my "best" presentations because the topics covered were diverse and seeing Data Scientists talk about their experiences was it's own reward. Quickly, it became clear that this is not just about statistics or machine learning. Our collective work is about working with people to reach goals that change society. This echoed the same sentiments expressed when I visited Data Science for Social Good Europe in June. Just look at these talk titles:

  • Using spatial features of human settlement to predict epidemic properties
  • Understanding maternal health service utilization
  • Machine learning for targeted communication in emergency
  • Crowd sourcing ‘Big’ clinical data from small health facilities
  • Data Revolution: A fitting Model for Development countries
  • Enabling Data Revolution
  • How Data Science is solving life-threatening problems in Africa plus the way forward
  • Understanding Africa's Wildlife Heritage Through the lens of Genome Data
  • Habari Node's Experience creating a Datacenter and Cloud Services Infrastructure
  • Mining voter sentiments from Twitter data for the 2016 Uganda Presidential elections
  • Algorithmic opportunities in revealing trends of food crisis from news online articles
  • Mobile Phone Data for Disasters Management
  • KAZNET: Leveraging digital and crowdsourcing technology for livestock market data collection
  • Sensing with Farmers; crowdsourced adhoc crop surveillance
  • A time series review of forest production and trade trends across the tropical region
  • Convolutional Neural Network for Appliance Recognition in Energy Disaggregation
  • Images - the all important developing world data format
  • Modeling Wireless Sensor Network For Forest Temperature and Relative Humidity Monitoring in Usambara Mountains - A review
  • A Weather Forecasting Model for Farmers in Arusha
  • Jaguza Livestock App
  • Air quality monitoring in Uganda
  • Bank At Hause – Factor Xchange
  • Monitoring economic indicators in Sub-Saharan Africa
  • Price predication for the agricultural commodities.
  • Prediction Modelling of Academic Performance, a Data Mining Approach
  • Challenges facing data management for community based education and services programs
  • Radio mining and rapid-deployment speech technology for humanitarian early warning in Uganda

The workshops should definitely be longer. There is just too much to do in 2 days. For that reason, 6/5?.

If you are thinking of attending Data Science Africa, definitely do!!! 😀

The other reason I was there

I got an opportunity to talk about the Deep Learning Indaba and our motivations putting together the almost weeklong Machine Learning tutorials/masterclasses in joburg. Luckily I was able to meet 2 participants from this years event and also chat at length to the other researchers about why we would like to collaborate and build on each others strengths. One thing that has become clear through the last 2 years has been how insular South Africa can be and also how disconnected we are. I stand to be corrected, but I was the only South African participant at the workshop. 5/5 ?.

Black in Artificial Intelligence

A number of us have been working together to increase representation in the field and there has formed a group (co-founded by Timnit Gebru) called "Black in AI". At Data Science Africa, 3 members were in attendance (including myself).


It was a quick trip to Arusha. But, because of using AirBnB I got a unique experience. My AirBnB was owned by the owner of Kitamu Coffee. Long story short, I love coffee and I had a lot of coffee and got a lot of coffee to take home with me 😀

Posted in CSIR, Data Science, Travel, Work Tagged with: , , ,

Teaching at AIMS Data Science Workshop

I had the privilege to be able to present a number of topics at the African Institute for Mathematical Science's Data Science workshop. I really enjoyed my time teaching, learning from the other speakers and meeting very talented students from around the country.

I presented 3 talks covering

  • Applications of Data Science to Social Good
  • Unsupervised Learning and Reinforcement Learning (Slides)
  • Visualisation (Slides)

Posted in CSIR, Data Science, Work Tagged with: , , , ,

Data Science #1: Science Stars Article

First article in the DST Magazine ScienceStars (Aimed at informing youth on science matters)

You can read it here.

Posted in CSIR, Data Science, Work Tagged with: , ,
Tweeter button Facebook button Linkedin button