Method of matching a set to evaluate and a reference list, corresponding matching engine and computer program

A method of indexing a computer file with tags from a reference list, the reference list being associated with a reference vector representative of the tags in the reference list, wherein the method comprises, in a matching engine comprising a hardware processor:

• associating textual content with the computer file;
• automatic extracting, from the textual content associated with the computer file, character strings and groups of character strings corresponding to keywords;
• calculating a distance between the reference vector and a vector associated with the computer file, the vector being representative of the keywords contained in the computer file;
• for each tag in the reference list:
• calculating a matching score for the keywords of the computer file and for the tag in the reference list, on the basis of the distance calculated between the reference vector and the vector associated with the computer file, the calculating of the matching score comprising:
• for at least one keyword contained in the computer file, calculating a centrality coefficient of the keyword, in the form of a sum of values of distance between the keyword and the other keywords of the vector associated with the computer file, weighted by a number of occurrences of the other keywords in the computer file;
• the matching score being calculated in the form of a weighted sum taking into account the distance calculated between the reference vector and the vector associated with the computer file and the centrality coefficient of the at least one keyword;
• calculating a distance between the vector associated with the computer file and a vector representative of constituents of the tags in the reference list, the calculating of a distance comprising:
• for each constituent of the tags in the reference list, calculating a matching coefficient for the constituent, in the form of a weighted sum taking into account the distance calculated between the vector associated with the computer file and the vector representative of constituents of the tags in the reference list, and the centrality coefficient of the at least one keyword;
• and
• for at least one tag in the reference list, calculating a second matching score for the computer file and the entry in the reference list, on the basis of the matching coefficients calculated for the constituents of the tag;
• providing a list of tags from the reference list, ordered according to the calculated matching scores and second matching scores;
• and
• automatic indexing of the computer file with tags in the ordered list of tags, without previous learning nor human intervention.