Al & Machine learningBig DataBlockchainCryptocurrencyData JournalismPredictive AnalyticsRobotics

Q&A with Data Scientist of the Week: Solomon Kimunyu

Solomon Kimunyu is a competent Data scientist, working in the Finance Department. He is currently ranked #1 out of over 370 data scientists who have enrolled in the cryptocurrency prediction challenge on Zindi. He was also ranked #5 out of 236 data scientists who participated in the AirQ Low-cost calibration challenge at Zindi. Solomon was ranked top 30% in machine learning worldwide after passing the machine learning Linkedin assessment and has participated in more than 10 competitions and hackathons among them: Hungry Geese competition – a reinforcement learning competition and the Bristol-Myers Squibb – a Molecular Translation competition.

Solomon has worked on a number of Data Science Projects including a Health recommender system, a Facial recognition system, and a pedestrian detection system. Last but not least, he is a data consultant at Pryce Machine Learning systems.

Predictive Analytics Lab has a  business intelligence app that allows you to make wise business decisions. Download the App, take photos, enter the details, then post and wait for it to reflect on the Location Intelligence Software.

1. What drew you to the Data Science Job?

Passion. I love working on data-related projects with my major interest being in machine learning, computer vision, and Natural language processing.

2. How did you acquire knowledge in Data Science?

Self-training and also enrolled in several courses on Udemy and Coursera. After completing the course, I practiced what I had learned by participating in Data Science related competitions and hackathons.

Predictive Analytics Lab formally launches a Partnership with the European Business University of Luxembourg which has given Scholarships in various Programs including Data Science and Blockchain with a total value of Kshs. 15 Million.

3.As a Data Scientist in your organization, what does the job entail?

I develop models which detect fraud in financial transactions, perform customer segmentation and predict customer churn. I also build models to perform sentiment analysis on the feedback that we receive from our customers and the comments that we receive from our clients via social media platforms. Aside from that, my job entails collecting data from various data sources such as databases, before preprocessing it and performing exploratory data analysis. Once done with that, I perform feature engineering and predictive modeling before working on hyperparameter optimization to improve the scores of the model. If the scores meet the threshold value, I compile a report on the whole process. But if the scores don’t meet the threshold, I will have to do more on feature engineering until I attain the desired results. Last but not least, I usually assist developers in deploying my models to the web.

4. How does a typical day at work look like?

My typical day starts at 7:30 am whereby I read through my Emails and respond accordingly. I then start working on my projects – currently working on Sentiment Analysis. The 1st thing that I do under this project is to collect data from Twitter and feedback on surveys that we share with our clients. After that, I preprocess the data, clean it then perform exploratory data analysis to understand more about the feedback that we receive from Twitter. Once done I classify the sentiments of the feedback into positive sentiments, negative sentiments, and neutral sentiments. I then try to understand the reasons as to why clients gave us that kind of feedback. After performing sentiment analysis, I write a report or do a presentation about the insights that I got.

A screenshot from the project presentation done by the Intermediate level students at Predictive Analytics Lab

In the afternoon, I help my team members to improve their models by testing the current model on the real data, evaluating the model, creating new training examples to fix the problem, and putting those models into production. The decision on when to stop improving the model and deploying it in production depends on the type of project.

The day ends at around 5:30 pm with a 30min of catch-up of the trends in artificial intelligence in Kaggle.

5. What makes you believe in the power of data?

It is a fact that data is increasing at an exponential rate which means that humans can extract more information from data than in the past. Computing power is also increasing and we can therefore process large chunks of data within a very short time.

6. What was the first data set you remember working with? What did you do with it?

I started off with the housing dataset where I used the dataset to fit a simple Linear Regression Model. I later did a Logistic Regression Model with iris and titanic datasets.

7. Looking back at your career in Data Science, what has been the most interesting project you have worked on? Why?

Fish classification, Cryptocurrency prediction, Medical imaging, and Speech recognition. I loved how my models worked.

Executives from 2 Financial Institutions in Ethiopia display their certificates after successfully completing a 5-day workshop on Digital Lending Product Strategy. The workshop was facilitated by the Predictive Analytics Lab in Nairobi Kenya.

8. What are the main types of problems now being addressed by Data Science?

Drilling information and useful insights from data.

9. What is that one thing that you have always wanted to do, but have never had a chance to do?

Creating a facial recognition system or a Biometric system for the banking industry and ATMs.

10. How would you advise someone interested in pursuing a career in the Data Science Industry?

Take some courses by Predictive Analytics Lab and then start practicing with real-world datasets.

Enroll in a Business Analytics and Data Science Course to start or progress your journey in Data Science. If you want to get constant updates and mentorship in the field, follow our Facebook, Twitter, Linked In and Instagram pages.

11. What is your parting shot?

Data has the solutions that we need to address the challenges that we are facing in the world.

For any Enquiries, You can drop us an email at sales@predictiveanalytics.co.ke or call /Whatsapp +254725349693

Tags

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Close

Adblock Detected

Please consider supporting us by disabling your ad blocker