Collaborative Research Assistant

download Collaborative Research Assistant

of 22

  • date post

  • Category


  • view

  • download


Embed Size (px)


Collaborative Research Assistant. 2007 Family History Technology Conference John Finlay Christopher Stolworthy Daniel Parker. Introduction. This presentation will introduce the Research Assistant module for PhpGedView It was developed by students from Neumont University - PowerPoint PPT Presentation

Transcript of Collaborative Research Assistant

  • Collaborative Research Assistant2007 Family History Technology Conference John Finlay Christopher Stolworthy Daniel Parker

  • IntroductionThis presentation will introduce the Research Assistant module for PhpGedViewIt was developed by students from Neumont UniversityTool designed to help genealogy researchersIdentify the problemsHow the Research Assistant help to solve those problems.Artificial Intelligence TechniquesResearch WorkflowHow the Research Assistant aids in the workflow

  • Identify The ProblemsTrack researchResearch is often duplicated due to inaccurate recordsResearch logs are not nearby when analyzing dataShare researchHow do I know what Uncle Bob in Ohio is researching?What has he already done?Determine what to researchIt can be difficult to analyze records and find the next thing to researchLosing placeIt is easy to forget where you were

  • Identify the ProblemsEnter resultsThere is a MAJOR GAP between the research results and the genealogy dataConsider the results of a census form and the wealth of data on itCurrently requires navigating through many, many different people and entering the same data over and over again

  • Identify the Problems6 people in the familyVerify names, relationship and genderAges give us approximate birth dates, birth placesOccupationsParents BirthplacesThe same source data entered up to 23 times!

  • Sharing & Tracking ResearchAll research is tracked through a Research TaskAssociated with multiple people/familiesKeeps a log of all research done for a personAssociated with a specific sourceLookup multiple research tasks at onceAssigned to a family member who will complete the taskKept with the genealogy data to simplify lookup and data entry

  • Analyze the data Analyze the data1Tracking ResearchResearch WorkflowDetermine possible sources ResearchEnter Results2345

  • Analyze the DataMissing InformationAnalyze a record and suggest missing informationAutomatically convert missing information into Research Tasks

    Nice, but how can we provide more?

  • Analyze the DataBayesian Data MiningArtificial Intelligence technique for predicting trends or highlighting anomalies in large data setsApplied to Genealogy we can use it to help predict events and places for researchersHelp researchers narrow and focus their effortsMost likely placeMost likely dateMost likely source

  • Analyze the DataCreate correlation rules of interestHow does a childs surname relate to his parents surnames?How does a childs birth relate to his parents birth?Use these rules to calculate probabilitiesEach dataset is uniqueDifferent cultures have different patronymicsSome groups tend to stay where they were born others where they were marriedCorrelation rules need to be uniquely calculated for different datasets

  • Analyze the Data

  • Analyze the DataLocal CorrelationsCalculate the rules with a smaller datasetLocalize the dataset around a person and their close relativesAverage the probabilities to get a more localized correlation

  • Analyze the DataWe can now apply these correlations to our missing informationSuggest the most likely places for events to occur

  • Analyze the DataFuture work to do:Possibility for AI to infer its own rules as it analyzes the dataCombine probabilities for rules that have matching dataWhat is the probability that the death place is Indiana given that the birth and marriage place are IndianaMore Bayes lawBroaden place localitiesCurrently only match on exact place matchBroaden to match on county and perhaps state

  • Tracking ResearchResearch Workflow Analyze the dataEnter Results145 Research3

  • Determining Possible SourcesHelp the researcher determine possible sources of their informationRequires a database of source information to look inExample to the right shows supplementing missing informa- tion with US census sources

  • Determining Possible SourcesFuture WorkImproved locality search. Again to broaden the search to match on county and state.Tie it into the FHL CatalogueCommon global repository for sources with a Web Service API we can query

  • Tracking ResearchResearch Workflow Analyze the dataEnter Results145Determine possible sources2

  • ResearchAuto-Search AssistantAutomatically pull data from a persons record so that it can be searched more easilyPluggable ArchitectureEasy to add new sites to searchDemonstration:http://localhost/pgv-nu/individual.php?pid=I6541&ged=test.ged&tab=5

  • Tracking ResearchResearch Workflow Analyze the data Research135Determine possible sources2

  • Entering ResultsUnique Source citation formsEnter in data the way it appears in the source recordEnter data only once!Structured forms allow us to automatically infer factsPluggable architecture allows us to easily add new forms

    Remember the 23 things to enter from the census record?Demonstrationhttp://localhost/pgv-nu/individual.php?pid=I716&tab=5

  • ConclusionPhpGedView Research Assistant Module simplifies technology for genealogy researchersAids in analyzing data through artificial intelligence techniquesHelps researchers find possible sourcesBrings research tools closer to the dataSimplifies data entryDistributed, Collaborative