As genealogists, we often struggle to narrow down our research focus and formulate clear objectives. Artificial intelligence (AI) can be a helpful tool in this process. Let’s explore how to use AI to find research questions and write objectives, using my Isabella Weatherford project as an example.
In our recent Research Like a Pro with AI Workshop, I followed the RLP process and explored ways AI could help streamline tedious tasks as well as provide new ways of approaching my research. In this six-part series, I’ll show each step of my research, beginning with the research question and objective.
Finding a Research Question
When starting a new project, it’s important to analyze your pedigree and identify gaps or areas that need further investigation. AI can assist in this process by analyzing your family tree data and suggesting potential research questions.
Using a GEDCOM and Custom GPT to Analyze Your Pedigree
Using a GEDCOM file with a custom GPT can provide a detailed analysis of your family tree. Here’s how to leverage this powerful combination:
Creating a GEDCOM File
First, export a GEDCOM file from your genealogy software:
1. Go to File > Export
2. Select the number of generations you want to include
3. Choose your starting person
4. Export only the basic facts, notes, and other relevant information
For my Isabella Weatherford project, I created a 10-generation ancestor GEDCOM, starting with my dad, Bobby Gene Shults.
Using a Custom GPT
Custom GPTs are specialized AI models trained on specific topics or tasks. For genealogy, you can use the Family Tree Expert custom GPT, which is free for all users (though you need a pro account to create your own). Custom GPTs are used via Open AI’s ChatGPT platform. On the left side of the home screen, click on “Explore GPTs.” You can search for “genealogy” or “family tree.”
Here are some tips for using custom GPTs:
– Ask the GPT what it can do
– Inquire about the best way to use it
– Explore its capabilities
Analyzing Your Pedigree with AI
Once you have your GEDCOM file and access to Family Tree Expert, try these prompts:
1. “Analyze this GEDCOM file and identify gaps or inconsistencies in the family tree.”
2. “Based on this GEDCOM, suggest four-generation research project ideas focusing on ancestors with events in [specific region].”
3. “Review this family tree data and propose research questions for ancestors with incomplete information.”
For my project, I refined the prompt to reflect specific ICAPGen Accreditation regions, such as the Upper South.
“Using this GEDCOM, suggest four-generation research project ideas where at least one life event occurred in Kentucky, North Carolina, Tennesee, Virginia, and West Virginia.”
The Family Tree Expert looked through my GEDCOM and found a possibility: the locations highlighted in the image below.
Benefits of Using a GEDCOM with Custom GPTs
1. Comprehensive analysis: AI can quickly process large amounts of family tree data
2. Identification of patterns: The AI might spot relationships or migration patterns you’ve overlooked
3. Tailored suggestions: Custom GPTs can provide research ideas specific to your family’s history and geography
4. Time-saving: Quickly generate multiple research ideas to explore further
Using an Ahnentafel Report to Analyze Your Pedigree
For the Isabella Weatherford project, I created an Ahnentafel report from Family Tree Maker for the same ten generations. I then uploaded this PDF to the AI tool, Claude 3.5 Sonnet. Here’s a prompt I used:
“You are an expert genealogist. Analyze this Ahnentafel report and suggest five research questions for Isabella Weatherford.”
The AI provided several interesting suggestions, including:
– What were the economic and social conditions in Dallas County, Texas, in the early 1870s that influenced Isabella’s life and marriages?
– How did Isabella’s migration from Missouri to Texas impact her life and family relationships?
– What factors led to Isabella’s two marriages in close succession (1874 and 1877)?
Tip: Refining Your Query
If the initial suggestions aren’t quite what you’re looking for, try refining your query. For example:
“Please suggest research questions focused specifically on Isabella’s early life in Missouri and her first marriage.”
Warning
The AI wants to please you. If you ask for something impossible in your pedigree, it will hallucinate and make up answers. For example, I asked Chat GPT 4.0 to find my ancestors who were born in Missouri between 1850 and 1900. In the image below, you can see highlighted the ancestors who did have a life event in Missouri but didn’t quite fit my parameters.
When I told the chatbot that this was incorrect, it explained that only Isabella D. Weatherford meets the criteria!
Using a Timeline to Generate Research Questions
Another way to give the AI tool information about a specific ancestor to review and find more potential research questions is to use a simple timeline.
Creating the Timeline
First, gather all the known facts about your ancestor. For Isabella Weatherford, I compiled information from various sources, including census records, marriage records, and her widow’s pension application. I organized this data into a simple timeline format:
1858, Mar 4 – Born in Missouri
1860, Sep 3 – Residing in Dallas County, Texas (Census)
1870 -unknown location
1874, Jan 16 – Married John W. Carpenter in Dallas County, Texas
1877, Jan 16 – Married Robert C. Royston in Decatur, Wise Co., Texas
1880 – Residing in Johnson County, Texas (Census)
1884 – Moved to Chickasaw Nation, Indian Territory
1915, May 2 – Death of husband Robert C. Royston in Duncan, Stephens Co., OK
1929, May 6 – Applied for widow’s pension in Oklahoma
1942, May 7 – Died in Oklahoma City, Oklahoma
An even easier way to create the timeline is the following:
- Go to a person’s profile in a family tree such as Ancestry or FamilySearch
- Highlight and copy their timeline data (CTRL +A, CTRL + C on a PC or Command +A, Command + V on a Mac)
- Go to an AI tool like Claude or ChatGPT
- Paste the timeline data and write your prompt, then hit “enter.”
Using AI to Analyze the Timeline
Once you have your timeline, you can use AI to help analyze it and generate research questions. Here’s a prompt I used with Claude:
“You are an expert genealogist. I would like to do a research project on Isabella Weatherford. Write a detailed narrative about Isabella from this timeline.”
After receiving the narrative, I followed up with:
“Now suggest five research questions from Isabella’s narrative.”
AI-Generated Research Questions
The AI provided several intriguing research questions based on the timeline:
1. What circumstances led to Isabella’s family’s move from Missouri to Texas between 1858 and 1860?
2. How did the Civil War impact Isabella’s family, given that she was a child in Texas during this period?
3. What factors influenced Isabella’s first marriage in 1874?
4. How did Isabella’s life change with her move to Indian Territory in 1884?
5. What was Isabella’s experience as a widow in early 20th century Oklahoma, and how did the Confederate widow’s pension affect her life?
I tried this exercise with both Claude and ChatGPT and received many suggestions. Finally, I settled on the following research question. Although I knew a fair amount about Isabella’s life after her second marriage, I knew nothing between 1860 and 1874. Despite my best efforts, I had not found the family on the 1870 census and didn’t even know where Isabella was living. I hoped learning more could shed light on this era of her life.
Isabella married twice before the age of 19 – first to John H. Carpenter in 1874 and then to Robert Cisnie Royston in 1877. What were the social and economic factors that might have influenced these early marriages?
Benefits of Using a Timeline for Question Generation
1. Identifies gaps: A timeline clearly shows periods where information is missing, suggesting areas for further research.
2. Highlights migrations: Movements between places often indicate significant life events worth investigating.
3. Reveals historical context: Placing an ancestor’s life events alongside historical events can lead to questions about how larger forces shaped their lives.
4. Spotlights inconsistencies: Contradictory information becomes more apparent in a timeline format, prompting questions to resolve these conflicts.
Writing a Research Objective
Once you’ve identified a research question, the next step is to formulate a clear objective. AI can help craft a well-structured objective that includes key identifiers.
For the Isabella Weatherford project, I used this prompt:
“Create a research objective for investigating the economic and social conditions in Dallas County, Texas, in the early 1870s and their influence on Isabella Weatherford’s life. Include key identifiers such as birth, marriage, and death dates and places.”
I gave the AI a sample objective so it knew how I would write it. The AI generated this objective:
“The objective of this research phase is to examine the economic and social conditions in Dallas County, Texas, in the early 1870s and their influence on Isabella D Weatherford’s life and marriage prospects. Isabella was born on 4 Mar 1858 in Missouri; she first married John H. Carpenter on 16 Jan 1874 in Dallas County, Texas, then later married Robert Cisnie Royston on 16 Jan 1877 in Van Zandt County, Texas, and died on 9 May 1942 in Tucumcari, Quay County, New Mexico.”
Now, I had an informative objective to guide my project!
Tips for Using AI with Research Objectives
1. Provide a sample objective with key identifiers to guide the AI
2. Ask the AI to reword your objective for clarity
3. Use AI to find new ideas for old, tired projects
4. Create a phased approach to your project by asking AI to suggest multiple objectives
How does this approach to using AI in your genealogy research sound? Have you tried similar methods? Share your experiences in the comments below!
Best of luck in all your genealogy endeavors!
Research Like a Pro with AI Series
Using AI to Find Research Questions and Write Objectives: Isabella Weatherford Project Part 1
Using AI in Timeline Creation and Source Analysis: Isabella Weatherford Project Part 2
Using AI in Locality Research: Isabella Weatherford Project Part 3
Using AI in Research Planning: Isabella Weatherford Project Part 4
Using AI in Research Logging: Isabella Weatherford Project Part 5
Using AI in Report Writing: Isabella Weatherford Project Part 6
Leave a Reply
Thanks for the note!