Narrative evaluator

A method comprising:

• receiving a plurality of narratives and an outcome for each narrative from a narrative repository, wherein the narrative repository is configured to store the plurality of narratives and, for each narrative, the corresponding outcome, wherein each narrative comprises a natural language description of characteristics associated with an event and the corresponding outcome for each narrative comprises an indication of whether the event was unapproved and required reporting to an administrative entity;
• for each received narrative, determining a subset of the narrative to retain based on predefined rules, wherein the predefined rules identify information to exclude from the narrative;
• for each determined subset, determining a factorized entropy matrix comprising for each word in the subset a set of numbers, wherein each number is associated with a count of co-occurrences of the word with another word in the natural language;
• for each factorized entropy matrix, determining an entropy matrix by multiplying the factorized entropy matrix by a transpose of the factorized entropy matrix, the entropy matrix comprising, for each word in the subset, a measure associated with whether the word is expected to appear in a sentence with another word in the subset;
• for each entropy matrix, determining a distance matrix by normalizing the entropy matrix using a partitioning coefficient, wherein the distance matrix comprises for each word in the subset, a numerical representation of a difference in the meaning of the word and another word;
• determining, using at least one of the distance matrices, for a first word of the subset, a first threshold distance, wherein, when a first narrative-to-word distance determined between a first narrative and the first word is less than the threshold distance, text in the first narrative is determined to be associated with a first outcome corresponding to an occurrence of an unapproved event, wherein the first narrative-to-word difference corresponds to a smallest difference between a meaning of the first word and a meaning of any word in the first narrative;
• and
• storing the first word and first threshold as a first word-threshold pair associated with the first outcome.

A device comprising:

• a memory configured to store a narrative repository, the narrative repository comprising a plurality of narratives and, for each narrative, a corresponding outcome, wherein each narrative comprises a natural language description of characteristics associated with an event and the corresponding outcome for each narrative comprises an indication of whether the event was unapproved and required reporting to an administrative entity;
• and
• a processor communicatively coupled to the memory, the processor configured to:
• receive the plurality of narratives and the outcome for each narrative;
• for each received narrative, determine a subset of the narrative to retain based on predefined rules, wherein the predefined rules identify information to exclude from the narrative;
• for each determined subset, determine a factorized entropy matrix comprising for each word in the subset a set of numbers, wherein each number is associated with a count of co-occurrences of the word with another word in the natural language;
• for each factorized entropy matrix, determine an entropy matrix by multiplying the factorized entropy matrix by a transpose of the factorized entropy matrix, the entropy matrix comprising, for each word in the subset, a measure associated with whether the word is expected to appear in a sentence with another word in the subset;
• for each entropy matrix, determine a distance matrix by normalizing the entropy matrix using a partitioning coefficient, wherein the distance matrix comprises for each word in the subset, a numerical representation of a difference in the meaning of the word and another word;
• determine, using at least one of the distance matrices, for a first word of the subset, a first threshold distance, wherein, when a first narrative-to-word distance determined between a first narrative and the first word is less than the threshold distance, text in the first narrative is determined to be associated with a first outcome corresponding to an occurrence of an unapproved event, wherein the first narrative-to-word difference corresponds to a smallest difference between a meaning of the first word and a meaning of any word in the first narrative;
• and
• store the first word and first threshold as a first word-threshold pair associated with the first outcome.