Find matching records

Sometimes your charts can contain multiple records that represent the same real-world objects and relationships. The Find Matching Records feature in Analyst's Notebook helps you to identify potential matches and reduce the number of matching records. This means you can keep key information together, uncover new connections and minimize duplication on your chart.

When multiple records exist for the same item, they are not necessarily populated with exactly the same information but do have some values in common. The likelihood of finding multiple records increases when you use data from a number of different sources, such as the Information Store and i2 Connect, and records that you create on the chart. For example, when you search sources for "Samuel Steele", you might find that records exist for "Sam Steele", "Sammy Steele", "S Steele", or other variations. You might have created your own records for "Samuel Steel" that include information that has not already been uploaded. It is possible that some or all of these records represent the same person and contain varying amounts of key information, source references, and notes. The match rules provided, along with the rules that you create, give you a fine level of control in searching out potential matches.

i2 Analyze records contain source identifiers that are used to identify the unique record in the source system that generated them. Although these source identifiers are not visible to analysts, Find Matching Records is able to match the source identifiers on records of the same type. Source identifiers can produce reliable matches because they are not subject to the possible inconsistencies present when property values are entered. For example, you might have two or more records for the same Person entity that are not discovered as a match, because the names are spelled differently. The rule provided for source identifiers can find these records, even when the property rules did not produce a match.

When relationships between entities are represented by multiple link records, your chart can become cluttered and difficult to read. For example, you might find matching links between Person entities and the property they have access to such as vehicles, addresses, or devices. You could improve the look of your chart by merging duplicate link records, as long as they link the same end record.

The presentation of the records in match sets allows you to easily compare the values that they have in common and examine any anomalies. You can then unite matching Information Store records behind a chosen target record on the chart, merge the information from chart records into a single target record, or resolve a combination of chart and Information Store records by both uniting and merging. For matching link records, the choice of which records to merge gives you the ability to control the links that you want to see on the chart, and those that you do not.