Hello everyone! I hope you all had a nice, sunny summer despite the rather upsetting situation at hand. Today, I am going to share one of the exciting things I learnt during the summer – coding in Python and its application in business!
Python for Business Intelligence is an introductory course in Python, part of the Information Systems specialization. During the summer, it was offered as an online course by Ashkan Fredström who specializes in the field of Entrepreneurship at Hanken’s Vaasa campus. Since it is a beginner’s course, it aims to familiarize students with Python syntax from a very basic level – variables, loops, if-statements, and exceptions, among others. This is taught through multiple exercises which are individual assignments.
On the flip side, the course is also well-suited for advanced users of Python as well. It familiarizes students with basic machine-learning and data-visualization techniques, a solid foundation to study more advanced topics that can be conveniently used in the thesis. Various business applications of Python are taught through group assignments. Overall, the course aims to enhance students problem-solving skills, which is one of the most sought-after skills in the corporate world.
An advantage of learning to code in Python is that it is an Object-Oriented Programming language which, in layman terms, means that objects are considered data structures containing their own customized functions. This allows programmers to reduce complexity in code and make changes without crashing the program. In the course, we learnt to make objects around a Point-of-sale System, with methods to add/remove products in a customer’s cart and a method to shop products, while on the back-end this acted as an inventory management system. Object orientation can also be adapted to applications in other fields such as Finance and Supply-Chain management – virtually speaking, the possibilities are endless.
Python boasts great integration capabilities with other programming languages such as C, C++, Java and even parse HTML using a library called BeautifulSoup. It features data structures such as tuples, dictionaries and lists which are particularly useful in interpreting data-interchange formats like XML and JSON. In the course, we learnt how to interface MongoDB, an online database which stores data in JSON, using Python commands. This can be used to conveniently store and retrieve records from online servers.
Another impressive feature is how fluently Python integrates various API services such as Google Maps, YouTube, and Twitter among many others. Along with extensive support libraries, this feature can be used to derive meaningful data that aids business decision-making. In the course, we learnt how to analyze comments from a YouTube video using polarity-based sentiment analysis. This was conducted through a Python library called VADER which uses natural language processing (NLP) techniques to gauge sentiments from social media text. Unlike more advanced NLP techniques, VADER is relatively simple to use and caters social media lingo like emojis and repetitive letters. Using this technique, for example, a business can judge how its latest video campaign is perceived on a scale of -1 to 1, and correspondingly make better decisions for the future.
In the last project of the course, we were introduced to machine-learning algorithms in regression modeling. The idea was to train the model on the ‘train data’, validate its parameters and then test its efficacy on the ‘test data’ with added hyper-parameterization. The project task was to predict the price of a car using a database of features and car prices. Machine-learning techniques such as Lasso, Decision Tree, Random Forest, Ada Boost and Gradient Boost were used to find the best performing model which had the highest overall R-square.
Python for Business Intelligence turned out to be one of the most enriching experiences of my journey at Hanken! Not only has it opened new avenues for my research, but it has given me a hands-on experience of how businesses can leverage Python to make data-driven decisions in the fast-paced world of today.
If you have any questions on studies or life at Hanken just let me know at firstname.lastname@example.org