Wanna skip the blabla and get right to the code? Access the Colab notebook here. I recently took a short course on DeepLearning.ai called Pair Programming with LLMs, where you learn how to use Google’s PaLM2 language model to help write, debug, and explain code within a Jupyter Notebook environment. Well, PaLM is old news, […]
Takeaways: ChatGPT can turn handwritten forms into data, even with sloppy handwriting. Defining a schema of the desired output helps. It makes mistakes. Output still needs to be validated and possibly fixed by hand. Can’t be automated with API yet. Still need to manually upload images to web application. Limit of four images per upload […]
One of the most annoying parts of data work is dealing with inconsistent entities: names of the same person spelled differently. Company names that rebranded, merged, or have varying suffixes like “Ltd.” and “Limited”. Standardizing data for accurate analysis can take days, sometime weeks, even with powerful tools like OpenRefine and Dedupe, which were made […]
TL;DR: Use the en_core_web_trf transformer model with Spacy to get much more accurate named entity recognition with multilingual text. Entity recognition is one of the marvels or current technology, as least from a journalist’s perspective. There was a time journalists had to read through hundreds, maybe thousands of documents, highlight names of people, companies and […]
One of the most exciting applications of AI in journalism is the creation of structured data from unstructured text. Government reports, legal documents, emails, memos… these are rich with content like names, organizations, dates, and prices. But to get them into a format that can be analyzed and counted, like a spreadsheet, usually involves days […]
And other insights from 7 years of anonymous Wikipedia edits by government employees
Datawrapper is right now the best tool for creating quick and simple charts. It’s so useful and feature-rich that news organizations that had their own in-house charting tool are switching over. One of its best features is the ability to connect a CSV file hosted on the web as a data source. This enables users […]
One of the final frontiers of data analysis is making sense of unstructured text like reports and open-ended responses in surveys. Natural language processing (NLP), with the help of AI, is making this kind of analysis more accessible. Libraries like spaCy and Gensim, although still code-based, are simplifying the process of getting insights out of […]
This is the text of the keynote speech I delivered at the 2018 Concordia Library Research Forum. It has been edited slightly. Thank you for this lovely opportunity to be among you this morning. I feel especially honoured to deliver this keynote because I believe librarians and journalists share a special kinship because our jobs […]
IMPORTANT UPDATE This post is outdated now that AWS Lambda allows users to create and distribute layers with all sorts of plugins and packages, including Selenium and chromedriver. This simplifies a lot of the process. Here’s a post on how to make such a layer. And here’s a list of useful pre-packaged layers. This post […]