When Visma students and AI work together
Find hidden treasures in the undergrounds of Jumbo-Visma’s Social Media using Sentiment Analysis.
Sentiment analysis, a prominent field in artificial intelligence, has grown in importance due to its ability to derive insights from human emotions and opinions expressed in text data. For the past few weeks, a group of enthusiastic and tech-savvy Visma students embarked on a sentiment analysis project that promised to be a transformative experience for both them and the communities they aimed to serve. In this article, we delve into the story of these passionate young minds as they collaborated on a project that brought together technology and human emotions.
First thing’s first: what is Sentiment Analysis?
Sentiment analysis is the process of analyzing digital text to determine if the emotional tone of the message is positive, negative, or neutral. Today, companies have large volumes of text data like emails, customer support chat transcripts, social media comments, and reviews. Sentiment analysis tools can scan this text to automatically determine the author’s attitude towards a topic.
The Vision
Our team of students shared a common vision: to create a sentiment analysis system that could understand and analyze emotions expressed during big events and based on Visma’s involvement in the cycling community, the focus landed on one of the world's most prestigious cycling events - the Tour de France. They recognized the potential of this technology to gain valuable insights from Jumbo-Visma’s social media posts.
The Implementation
-
Armed with determination, the students began the implementation phase of their sentiment analysis projects. Leveraging open-source machine learning libraries, they designed a system that could analyze vast amounts of textual data in real-time. The sentiment analysis model was trained on different Jumbo-Visma accounts: Instagram, Youtube, Facebook, Twitter.
-
Based on the project our students mostly used Python (high-level, general-purpose programming language, PIP (is a package-management system written in Python and is used to install and manage software packages) and sourced different libraries. From an AI perspective, Instagrapi (unofficial Instagram API for Python) and OpenAI (Chat-GPT).
Empowering the Future
As the internship came to a close, the students reflected on their incredible journey. They had not only successfully completed a sentiment analysis project but also experienced personal growth and discovery. But let’s hear it directly from one of their mentors, Radu Bălan, team manager for the 24/7 team in Visma.
-
Over the past few weeks I have had the pleasure of working and mentoring this diverse group of talented students. During their practice with us, several students exhibited remarkable dedication, enthusiasm, and a genuine passion for their work. Their thirst for knowledge and eagerness to contribute to our company's projects were evident from day one. These interns were not only highly interested in their assigned activities but also took the initiative to seek out additional learning opportunities within the organization.
Radu Bălan
Team Manager
A stepping stone towards AI-driven insights
The sentiment analysis project served as a stepping stone towards a future where AI-driven insights empower decision-making and positively impact lives, but don’t take our word for it, let’s hear it from some of our students:
-
I had the opportunity to work within Visma for a period of 3 weeks on a project regarding NLP (Natural Language Processing) on Le Tour de France using Sentiment Analysis. My work started with a Java project using the Stanford CoreNLP API, but I soon realized it was not the best direction. After some discussions with my mentors I made a judgment call: the better approach was to use the OpenAI API with Python. The platform I was tasked to collect data from was Instagram and, after that, provide a visual representation of everything gathered. Lucky to have mentors that provide useful insight and helped us deliver polished projects. I am confident that the skills and insights we all gained in Visma will serve as a strong foundation for future endeavors.
Pitea Emil
Visma student
-
When I was accepted as a practice student at Visma, I had mixed feelings about this: on one hand, I was thrilled to see how such a big company runs, but on the other I was concerned if I am going to meet the expectations. My doubts disappeared in a matter of seconds because of the people who worked with us. They made me feel like I am part of the company. After 3 weeks of practice, this proved to be such an amazing journey and experience that no one can take away from me.
Rosca Sonia
Visma student
-
When I started this summer practice, it felt like a big challenge, but it also felt like a fulfilled dream of mine when I heard that I would be working with elements related to Machine Learning. It was a starting point for my future career, and I’m so thankful for this opportunity. All the people there helped me grow and never lose interest in this fascinating topic of AI that we covered. I still need to work on my journey, but I’ve gained a lot of important knowledge to carry on my back and some memories that will never be forgotten.
Vlăduț Cosmin
Visma student
Conclusions
The students' sentiment analysis project during the Tour de France was an exemplary display of the potential of technology and human collaboration. It showcased how a passion for AI and a vision for change can lead to innovative solutions with far-reaching consequences. Through their journey, these students proved that the synergy of technology and human emotions can create powerful insights, shaping a better, more informed world. As they venture forth, their story inspires countless others to embark on similar journeys, exploring the uncharted territories of AI and its transformative capabilities.
P.S. No AI was used in writing parts of this article - xoxo ChatCPT.