Pattern Recognition as a Goal in Genealogy

Quoting from the Journals.Elsevier.com website about the journal, "Pattern Recognition."

Pattern Recognition is a mature but exciting and fast developing field, which underpins developments in cognate fields such as computer vision, image processing, text and document analysis and neural networks. It is closely akin to machine learning, and also finds applications in fast emerging areas such as biometrics, bioinformatics, multimedia data analysis and most recently data science. The journal Pattern Recognition was established some 50 years ago, as the field emerged in the early years of computer science. Over the intervening years it has expanded considerably.

Genealogy is really all about patterns. This fact was recognized many years ago when genealogists began applying what is known as "cluster research" to their research. Cluster research consists of adding research to the extended family, friends, and neighbors of ancestors. But pattern recognition is much more than cluster research. If genealogists were to use pattern recognition, they would have to add research into the historical, cultural, social, religious, and occupational background of their families.

I ran into an issue of pattern recognition recently when I discovered a family line living in the county of Kent, England. So far, for this family, the predominant occupation is that that of basketmaker. It appears that basketry extends back generations. In doing some research about individuals who were basketmakers, I found that traditionally, this occupation was practiced by those of Romi or Romany descent. Now, I am beginning to see a pattern and even when the individuals move to other areas of England, I can see how their trade of basket making has been preserved. This particular pattern begins to disappear during the development of the industrial revolution in England.

Failure to view genealogical research as based on reconstructing family patterns is probably one of the most limiting factors in genealogy's acceptance as an important academic pursuit.

To begin using basic pattern recognition techniques to advance genealogical research, there needs to be a greater emphasis on placing the family within the context of its entire background. The most basic pattern is that of where events in a particular family occurred. The current emphasis in genealogical research is on adequate documentation of the family relationships. This goal is usually referred to as conducting a reasonably exhaustive research effort. However, simply looking for documents and other sources about the family begs the issue of actually identifying the family in the context of its existence as a patterned entity. Granted, there are some very competent genealogical researchers that have knowledge of the history and backgrounds of their families, but the most genealogists concentrate only on "genealogically significant" records.

Take for example a core family unit. Looking at a family as merely a biological unit is tantamount to looking at the world in black and white rather than color. There is amazing detail in black and white photographs but there is a whole dimension of additional detail given when the photos are in color.

When we move into the realm of online, digitally based genealogical research, it is easy to stay at the searching for names level of genealogical research. Moving beyond that level may actually involve reading a book or visiting the area where your ancestors lived. It may also involve a lifetime learning process of gaining sufficient knowledge about the history, geography, and culture of the places where your ancestors lived to reach the level where you know who they were and who their ancestors were.

I am only beginning to recognize the patterns of my own ancestors after more than 36 years of research. But the benefit of what I have learned helps me to do original research for unknown family members with more accuracy than I could have believed possible even a few years ago.