- Design for Data Visualisation, and Introduction to Maps and Spatial Data
AIR - Academy of Innovation & Research
This meet-up will cover two themes: The presentation of data as information is a discipline that brings together design, ethics, statistics and the psychology of perception. Geographic maps are not just a familiar and efficient form of visual communication - they have a history rich in human stories, sometimes happy, and sometimes urgently highlighting human suffering. --- We're very lucky to have Caroline Robinson, BA ACMI FRGS Founder and Lead Cartographer at Clear Mapping Company: * introduce us to the key design principles for data visualisation, * also lead a tutorial on working with spatial data and using open source geographical information system (GIS). After this session you'll be able to design better data visualisations, and have the confidence to use spatial data and GIS tools for your own projects. --- 40mins - short talk on design principles and foundation of visual communication, covering colour, fonts, culture and equality in data collection - Make your #maps sing! 40 mins - hands-on workshop creating and using spatial data, learning how to make our own points, lines and polygon data, and how to export the data for your web based projects. --- The session will be suitable for complete beginners - you don't need any previous experience in design, data visualisation, mapping or using geographical information systems. A colourful and insightful evening into the world of spatial data! --- We'll start at 7.00pm, and you can arrive from 6.30pm. Parking is free after 5pm. Bring a fully charged laptop with working wifi if you'd like to follow the hands-on tutorial. A scroll-wheel mouse is recommended if your trackpad isn't good for scrolling/zooming. * image from: https://geoffboeing.com/2018/07/city-street-orientations-world/
- Python First Steps - A Hands-On Tutorial
A beginner-friendly hands-on introduction to coding with Python. ---- Python is one of the most popular and broadly used programming languages*. You'll find it at the heart of global-scale technology services and used by children to control small computers and robots. Python has also become the de facto standard language for data science - used for data processing, analysis, machine learning, and visualisation. ---- This session is a hands-on introduction to Python. No previous experience of Python, or any coding, is necessary. If you've never coded before, you're welcome to join us for a supportive and friendly session to learn with others also starting their journey. We'll start at zero and work through the basics - using the python web-based notebook, variables, loops, functions .. and the moving onto useful things for data scientists - importing useful libraries, working with data frames, simple visualisation, saving and loading data. Don't worry if none of these words mean anything yet! By the end of the session you'll have written your own code, make a good start at reading and learning from existing Python code, be able to use the very vibrant ecosystem of libraries .. and gained the confidence to continue learning! --- Bring a fully charged laptop with working wifi. There's no need to install any software as we'll work entirely on the web. You'll need a Google account. If you don't want to create a Google account, you can install the Anaconda Python and use that - help will be provided. --- We'll start at 7.00pm but you can arrive from 6.30pm. Car parking is free from 5pm. --- * the Economist discusses the rise of Python: https://www.economist.com/graphic-detail/2018/07/26/python-is-becoming-the-worlds-most-popular-coding-language
- Machine Learning for Image Classification - Hands-On Tutorial with Tensorflow
AIR - Academy of Innovation & Research
A beginner-friendly hands-on image machine learning for classification tutorial with Tensorflow. -- In recent years there has been huge progress in machine vision and image classification, stuff previously only dreamt of in science fiction. We're lucky to have Barney Nicholls lead a beginner-friendly tutorial on image classification using Tensorflow. He'll briefly set out a business challenge that inspired him to look at image classification, before leading us gently through a simple worked example of training a Google Tensorflow neural network to recognise images, and then testing it. He'll weave in wisdom and insights into machine learning as we progress through the example. He'll also provide suggestions for how the simple solution could be improved, and point to modern alternative approaches to Tensorflow. This hands-on tutorial is for beginners and you don't need to have any previous experience with Tensorflow, machine learning or image classification. -- David T will also do a 5 minute flash talk at the beginning explaining some of his work on sentiment analysis inspired by our previous meetup. -- Please bring a fully charged laptop with working wifi. We'll be using cloud services so you shouldn't need to install any additional software.
- Sentiment Analysis - Hands-On Tutorial with Python
Natural language is everywhere - from legal documents to tweets, from corporate emails to historic literature, from customer discussions to public inquiry reports. The ability to automatically extract insight from text is a powerful one. The challenge is that human language is hard to compute with as it was never designed to be consistent, precise and unambiguous - in fact, that's its beauty! We're very lucky to have Peter Conway of Hertzian lead a hands-on tutorial on simple natural language processing and sentiment analysis using Python - the leading programming language for data science. The tutorial is designed for newcomers to natural language processing (NLP) and is suitable for beginner programmers. Peter will introduce just enough Python as the tutorial proceeds. By the end of the tutorial you'll be able to analyse text for sentiment, as well as have a basic understanding of how sentiment analysis works. Bring a fully charged laptop with working wifi. You'll need Python installed with some common data science libraries installed - Anaconda Python is the recommended and simple way of doing this. https://www.anaconda.com/download/ The tutorial will start at 7.00pm. You can arrive from 6.30pm and the opportunity to meet other members of our community.
- How To Design for Big Data
A successful product will often see a growth in the data it stores and processes. The downside of this success is that data storage can grow beyond the limits of a single traditional database, and data processing can outgrow the memory limits of a single computer. We're very lucky to have someone as experienced as Rob Harrison to talk us through: 1. The common problems that emerge as your data outgrows your application. 2. The key design principles for architecture that can scale beyond a single-computer database and memory. 3. Examples of technology and designs that have proven themselves to scale - including at CERN. Rob will explore: * Designing for success - architecting redundant applications which continue to perform under load. * Parallelism and programming paradigms - writing applications that allow for both fast iteration and optimisation. * Benefits of microservices - how to perform distributed processing of large data sets across clusters. Whether you're starting small, starting big, or growing an existing service, this talk will provide valuable advice for successfully scaling your products and services - advice based on Rob's practical experience of what works and what doesn't. --- We will also have two short flash talks: 1. John Moore explaining his work on AI generated game narratives, and inviting participation in his project. 2. Ian Mason from Exeter University on the Smartline for-social-good research project exploring sensor data from homes in Cornwall, and an invitation to participate in data mining hackathons. --- We will also have a flash talk on using AI to develop narratives and storylines for games.
- Python for Data Science - Top 10 Tools
Python is a very popular language that has also become the de facto tool for data science. In this beginner-friendly overview, we'll explore the most popular Python libraries used to do data science. We'll look at tools for data storage and processing, machine learning, natural language text processing, visualisation and sharing your work with notebooks. We'll also look at very brief examples to show these tools in action. We also hope to have a short 5 min talk about using AI to generate multi-threaded storylines for games. ---- The event is 18.30 to[masked] networking[masked] welcome and main talk[masked] networking / discussion Parking is free if you arrive after 17.00