Abstract: A system for managing genealogy data is provided. The system includes a genealogy application for storing genealogy account identification data for a first genealogy account. A social network interface receives data from the first genealogy account and provides genealogical data to the genealogy application. A relative invitation system generates an invitation to a user to create a second genealogy account for the genealogy application.

Abstract: Record Linkage (RL) is the task of identifying two or more records referring to the same entity (e.g., a person, a company, etc.). RL models can be based on Cost Sensitive Alternating Decision Trees (ADTree), an algorithm that uniquely combines boosting and decision trees algorithms to create shorter and easier-to-interpret linking rules. These models can be naturally trained to operate at industrial precision/recall operating points, and the shorter output rules are so clear that it can effectively explain its decisions to non-technical users via score aggregation or visualization. The models significantly outperform other baselines on the desired industrial operating points, and the improved understanding of the model's decisions led to faster debugging and feature development cycles.

Abstract: Records are obtained from various sources, including public records, and are then augmented and grouped. At least one augment key is applied to each of the records resulting in an augment key value for each record. The records are augmented based on the key value. Augmenting the records includes combining the fields of the common records into a single record and then removing the duplicate records. The augmented records are then grouped according to household. At least one household grouping key is applied to each of the augmented records resulting in a household grouping key value for each record. The records having the same household grouping key value are displayed as a household grouping.

Abstract: Dynamic blocking determines which pairs of records in a data set should be examined as potential duplicates. Records are grouped together into blocks by shared properties that are indicators of duplication. Blocks that are too large to be efficiently processed are further subdivided by other properties chosen in a data-driven way. We demonstrate the viability of this algorithm for large data sets. We have scaled this system up to work on billions of records on an 80 node Hadoop cluster.

Abstract: Dynamic blocking determines which pairs of records in a data set should be examined as potential duplicates. Records are grouped together into blocks by shared properties that are indicators of duplication. Blocks that are too large to be efficiently processed are further subdivided by other properties chosen in a data-driven way. We demonstrate the viability of this algorithm for large data sets. We have scaled this system up to work on billions of records on an 80 node Hadoop cluster.

Abstract: Record Linkage (RL) is the task of identifying two or more records referring to the same entity (e.g., a person, a company, etc.). RL models can be based on Cost Sensitive Alternating Decision Trees (ADTree), an algorithm that uniquely combines boosting and decision trees algorithms to create shorter and easier-to-interpret linking rules. These models can be naturally trained to operate at industrial precision/recall operating points, and the shorter output rules are so clear that it can effectively explain its decisions to non-technical users via score aggregation or visualization. The models significantly outperform other baselines on the desired industrial operating points, and the improved understanding of the model's decisions led to faster debugging and feature development cycles.