In an approach for comparing items of anonymized data, a processor receives a plurality of items of data, each item comprising an anonymized ordered list of words, wherein each word is anonymized in one or more forms and is associated with a respective item of data. A processor compares a first set and a second set of the plurality of items of data by: comparing each of the one or more forms of each word in the first set with the corresponding form of each corresponding word in the second set, scoring each comparison of each word based on a degree of matching between each of the one or more forms of each word in the first set with the corresponding form of each corresponding word in the second set, and computing a total score for the comparison of the first item and the second item based on the scoring.