Handwritten text recognition by large language models (LLMs) has become much better with the latest releases of ChatGPT 4o and Claude 3. 5 Sonnet. Claude 3.5 Sonnet is the smartest LLM available today. See Anita Kirkovska, “Claude 3.5 Sonnet vs GPT-4o,” vellum (https://www.vellum.ai/blog/claude-3-5-sonnet-vs-gpt4o). I decided to test Claude 3.5 Sonnet with one of my favorite tasks – handwritten text recognition, a challenging task for LLMs. To test the capabilities of Claude 3.5 Sonnet, I created a...
In this podcast episode, Diana and Nicole discuss the use of Generative AI in scholarly and genealogical writing, emphasizing the importance of transparency in disclosing AI assistance. They talk about editorial guidelines from scholarly journals and the Association of Computational Linguistics, which suggest clear declarations of AI’s involvement in literature searches, drafting, and idea generation. Key points include recommendations for crediting AI-generated content not as authors but by detailing the AI’s role. They also explore...
If you have colonial American ancestors, you may or may not have been successful in your quest to find the original immigrants. This can be difficult for a number of reasons. Because people from Britain were emigrating to the British colonies, not many records were kept. Also, the early years in the Americas saw colonists struggling to survive in this new land. Record keeping would not have been high on the list of priorities...
In this podcast episode, Diana and Nicole discuss the evolving role of AI chatbots in various writing contexts, particularly in academia and genealogy. They highlight the changing policies around the use of large language models (LLMs) in academic writing, noting how organizations like the International Conference on Machine Learning have shifted from prohibiting to allowing LLMs, provided that authors are responsible for the content and its integrity. They delve into the ethical considerations of using...
In part 1 of this series, we discussed records of the parish chest that can help our research. This blog post will feature tax records, a valuable resource beyond parish collections. One might think taxes aren’t helpful to genealogy since they only name the individual paying the tax: typically, the head of a household who met the tax’s qualifications (age, residence, ownership of property, etc.). But tax records are invaluable because they date much further...
In this episode of “Research Like a Pro,” guests Elaine Martzen and Emmaline MacBeath join hosts Diana and Nicole to discuss various WikiTree projects. Projects on WikiTree are groups organized around topics or activities to facilitate collaboration among members. They include top-level projects with unique tags and badges, sub-projects using higher-level project badges, project teams, and informal free-space projects. Types of projects include geographical (e.g., Australia, Mexico, South Africa), themed (e.g., Mayflower, Holocaust, Native Americans),...
Anyone researching their English ancestry knows how important parish records are. The British government began documenting births, marriages, and deaths in 1837, but before that, it was up to the Church of England to record the life events of its parishioners, typically in the form of baptisms, marriages, and burials. These records are essential for proving pre-1837 relationships and tracing your family back through time. But what if you simply can’t find your ancestor’s baptism,...
This episode of Research Like a Pro focuses on the DNA features available on Wikitree, with guest Emmaline MacBeath, a genetic genealogist, explaining them. Diana and Nicole, discuss with Emmaline the various DNA tools and functionalities on Wikitree. One key feature is the ability to mark relationships as “DNA confirmed” by providing evidence from DNA matches. Users can also record the DNA tests they’ve taken on their profiles, which propagates to ancestors and descendants within...
When we have a female brick wall ancestor with no clues to her origins besides a maiden name, careful research can provide a hypothesis that DNA evidence can confirm. In the case of my third great-grandmother, Clemsy (Cline) Weatherford, the research was challenging because of record loss in the northeastern Arkansas counties that could have held records stating relationships. I discussed the documentary research in federal and state records in my post, 5 Tips for...
If you’ve tried researching in an area with severe record loss, you’ll know this comes with many challenges. Often, the records you need most to prove relationships don’t exist: probate, marriage, deeds, etc. All is not lost, though; with perseverance, you can make progress one step at a time. Using my research in northeastern Arkansas burned counties, I’ll show examples of how I reconstructed a family using alternate jurisdictions, attention to detail, and correlating every...
