MachineHack successfully concluded its 15th instalment of the weekend hackathon series last Monday. The Melanoma Tumor Size Prediction: Weekend Hackathon #15 hackathon was greatly welcomed by data science enthusiasts and practitioners.
Out of the 205 competitors, three topped our leaderboard. In this article, we will introduce you to the winners and describe the approach they took to solve the problem.
#1: Devesh Darshan
Currently, in his second year of Engineering at Birla Institute of Technology and Science, Pilani, Devesh first came across the term data science during his first year. Like many, he started his journey with the popular Stanford University course by Andrew Ng. His curiosity led him to many other popular online courses as well. He started practising with simple data sets like Titanic and House price on Kaggle. He spends most of his time reading articles and blogs of Analytics India Magazine and Medium to learn new ML techniques.
Approach To Solving The Problem
Devesh explains his approach as follows:
The problem seemed simple at first site, but later I realized that to get a good score in this competition, you just couldn’t rely only on modelling and different ensembles. The things that helped me to reach a good score were:
- Feature Engineering: This played a major role for me. Engineering relevant features from the dataset was a challenge at first, but then I was able to create 13 new features that helped the model a lot.
- Model Selection: Surprisingly, Extra Trees Regressor worked the best here, so choosing the right model was also very important in this problem.
- Ensembling: The final edge was provided by Stack Ensemble of 5 models that were least correlated with each other.
“MachineHack is the best platform for new data scientists to practice and test their skills, as some of the problems stated are very beginner-friendly, unlike Kaggle or other platforms where the problems require experience and a better machine to implement the solution” – Devesh shared his experience.
#2: Harshita Gupta
Harshita is a third-year student currently pursuing Civil Engineering at Birla Institute of Technology and Science Pilani. She started her data science and machine learning journey in her second year at college by taking some relevant courses from the institute. To get a better insight into the field she did various online courses as well. She also prefers reading a lot of blogs and articles.
She constantly participates in hackathons to improve her practical knowledge in the domain.
Approach To Solving The Problem
Harshita explains her approach as follows:
The approach was really simple, I started with :
- Exploration of data using correlation matrix, data analysis and distribution of the features
- Feature engineering, as it is an important step that could significantly improve my score
- Finally trying out different regressors such as LGBMRegressor, BaggingRegressor, ExtraTreeRegrssor, XGBoostRegressor and RandomForestRegressor.
- The major step which boosted my score was using ensembles by combining different regression models and fine-tuning.
“MachineHack is an amazing platform for beginners. I really enjoy applying my theoretical knowledge on their hackathons conducted every weekend.Due to a huge competition on the leaderboard, one is really pushed to the limits to solve the problem, thus encouraging a person to improve a lot. Glad to be a part of the MachineHack community” – she shared her opinion
#3| Devrup Banerjee
Although Devrup learnt python just out of the sheer need to automate the routine work and gather data at scale, his real enthusiasm and passion for data science sprouted in his second year of MBA at Great Lakes Institute of Management, Gurgaon, while he was attending his marketing and retail analytics class. He realised that the real motivation behind learning all these algorithms was not about enhancing accuracy but to tell your client by how much you can promise to increase their bottom-line if they were to follow your exact given path. The subject changed his life.
“My roommate, who was also equally inspired, and I used to have sleepless nights just going through the 25 lacs dataset given as a final project with our rickety computers to generate actionable insights. To better the bottom-line percentage, that’s what inspired me into analytics.“ – He said
His team has won many competitions at MBA level, won at IIT Kanpur MRA tournament while finishing as runners up at IIM Kashipur’s case study on analytics.
He is currently trying to deep dive into data science to better his analytical skills so that if someone gives him a dataset in future, he can be both a business analyst and a data scientist.
Approach To Solving The Problem
Devrup explains his approach briefly as follows:
The train and test sets contained lots of duplicates. Some of those duplicates in the test set exactly matched those within the training data. So I grouped the values in the train by mean of target and manually imputed all the exact matches with this mean in the test. There were a couple of zero values which needed to be handled too. Then came the blending of two relatively strong yet different model setup, combining which resulted in good scores.
“Machinehack is doing a great job organizing the hackathons and data science summits which not only provides exposure to budding data scientists and students like me, but also helps in opening up networking opportunities with the very best in the domain”- he shared his opinion about MachineHack
