PyData Bristol - 26th Meetup (New Date)

# Details

Join us once again for the next PyData Bristol Meetup!
Sorry everyone for the big delay from our last meetup. We’re meeting now in a new venue - Ovo!!! Big thank you for hosting our event this time. We are saying goodbye to our long lasting event host - Cookpad, thank you for being with us for the last year . Massive thanks to our other sponsor - Adlib for the scrumptious pizza, and thirst-quenching refreshments.
Agenda for the evening:
🚪 6:00 pm - Doors open
🕡 6:30 pm - Talks commence (sharp!)
📚 1 25-minute talks:

• Talk 1: Natural Language Processing - from Academic Theory to Business Application by Jerry Panashe Mundondo

📚 Two 10-minute lightning talks:

• TweetNLP by Asahi Ushio
• TBD

📢 Community announcements
🤝 Relaxed networking over beers and soft drinks
Interested in sharing your knowledge or experience at this or a future event? Fill out this form to submit your talk proposal: PyData Bristol Talk Proposal

We look forward to seeing you there for another fantastic evening of Python, Data Science, and camaraderie!
‌
‌

# Talks

## 1. Natural Language Processing - from Academic Theory to Business Application - Jerry Panashe Mundondo

The speaker will center their talk around the contrast between academic data science, as experienced during their MSc in Data Science & AI, and the adaptability required in real-world data scenarios.
Using an NLP tool they developed for Fourth Floor Creative (a video game marketing company) as a case study, they will discuss the challenges in adapting from data science in theory to the creativity and solution-oriented approach needed in the real world.
The aim of the NLP tool was to evaluate both sentiment and semantic themes in customer comments about specific video games. Initial attempts using off-the-shelf models like TextBlob and Bert were not accurate enough due to the unique language in the gaming domain.
Holding onto the academic mindset, they contemplated creating a bespoke sentiment classifier, only to discover the lack of clean, labeled data specific to their domain—a stark contrast to academic settings.
Ultimately, rather than building a tool from scratch, the solution lay in fine-tuning existing models from Hugging Face, augmenting data with synthetic inputs from LLMs, and capitalizing on few-shot classification to address data limitations.
The speaker believes that early-career data scientists will benefit from this talk. It provides insights into the practical application of data science and NLP in a business context, highlighting the emphasis on speed and efficiency, as opposed to the academic focus on achieving the highest accuracy in a model.
‌

# Lightning Talks

## 1. TweetNLP - Asahi Ushio

Language models have been widely used in many natural language processing (NLP) applications, but they do not perform well on social media due to the noisiness and temporal nature. In this talk, we will shortly describe our studies of NLP on social media, and introduce TweetNLP, a python library to facilitate the use of NLP tools tailored to social media.

## 2. TBD

‌
🕖 LOGISTICS
Talks kick off at 18:30 sharp; then networking in The Knights Templar from 20:40.
If you realise you can't make it, please un-RSVP in good time to free up your place for your fellow community members.
Follow @‌pydatabristol (PyData Bristol (@PyDataBristol) / X) for updates on this and future events, as well as news from the global PyData community.

📜 CODE OF CONDUCT
The PyData Code of Conduct governs this meetup (Code of Conduct
). To discuss any issues or concerns relating to the code of conduct or behaviour of anyone at the PyData meetup, please contact the PyData Bristol organisers, or you can submit a report of any potential Code of Conduct violation directly to NumFOCUS (NumFOCUS Code of Conduct Report Form).