Classification code parser for identifying a classification code to a text
A classification code parser and method can include: reading a classification code having a description; reading a required keyword, and a total number of keywords associated with the classification code; reading text of a note; tokenizing the text of the note to create a note token stream, the note token stream having a note token and a position of the note token within the note token stream; creating a keyword map including a total number of matched keywords; determining a match ratio from the total number of the matched keywords and the total number of the keywords; determining a proximity factor based on a shortest span of tokens within the note token stream containing all the matched keywords; and determining a strength of a match between the classification code and the note based on the match ratio being multiplied by the proximity factor.
1. A method comprising:
obtaining a classification code and a text;
generating, from the text, a note token stream, wherein the note token stream includes a note token and a position of the note token, wherein generating the note token stream comprises: removing negative content identified from the text, the negative content comprising a portion of the text starting with a negative term;
generating a keyword map including a keyword associated with the position based on the keyword matching the note token, and the keyword map including a total number of matched keywords associated with the classification code;
determining a match ratio from the total number of the matched keywords and the total number of the keywords;
determining a proximity factor based on a keyword span; and
determining a strength of a match between the classification code and the text based on the match ratio and the proximity factor.
2. The method of claim 1 , wherein the classification code includes an ICD-O-3 code and a behavior code.
3. The method of claim 2 , comprising determining a behavior factor according to the following equation:
if the behavior code is “3” or if the behavior code is “2” and the text includes “non-invasive”, then the behavior factor is 1.5;
otherwise, the behavior factor is 1.0; and wherein determining the strength includes multiplying the match ratio, the proximity factor, and the behavior factor.
4. The method of claim 1 , wherein determining the match ratio includes determining the match ratio according to the following equation:
match ratio=k 2 /m, where k is the total number of the matched keywords and where m is the total number of the keywords.
5. The method of claim 1 , wherein determining the proximity factor includes determining the proximity factor according to the following equation:
proximity factor=1/(1+e/50), where e is a largest value between (s-m) and 0, where s is the shortest span of tokens within the note token stream including the matched keywords, and where m is a total number of the keywords in the classification code.
6. The method of claim 1 , wherein generating the keyword map comprises:
reading a secondary keyword and an ignored token;
determining a primary keyword as any of the keywords that are not a required keyword, the secondary keyword, or the ignored token; and
determining whether the primary keyword matches a second note token within the note token stream.
7. The method of claim 1 , wherein generating the note token stream further includes:
reading a special token; and
converting the text to lowercase unless the text matches the special token.
8. The method of claim 1 , wherein generating the note token stream further includes:
obtaining a descriptive synonym; and
replacing the note token with the descriptive synonym based on the note token matching the descriptive synonym.
9. The method of claim 1 , wherein generating the note token stream further includes:
splitting the text into a plurality of sentences;
identifying a negative term from one of the plurality sentences; and
deleting a portion of the sentence of the plurality of the sentences from the negative term to an end of the sentence.
10. The method of claim 1 , wherein the keyword span is the shortest span of tokens within the note token stream including the matched keywords.
11. The method of claim 1 , wherein determining the strength includes multiplying the match ratio and the proximity factor.
12. The method of claim 1 , comprising tokenizing the text of the note to create the note token stream.
13. A non-transitory computer readable medium storing instructions that, when executed, cause one or more processors to:
obtain a classification code and a text;
generate, from the text, a note token stream, wherein the note token stream includes a note token and a position of the note token, wherein generating the note token stream comprises:
removing negative content identified from the text, the negative content comprising a portion of the text starting with a negative term;
generate a keyword map including a keyword associated with the position based on the keyword matching the note token, and the keyword map including a total number of matched keywords associated with the classification code;
determine a match ratio from the total number of the matched keywords and the total number of the keywords;
determine a proximity factor based on a keyword span; and
determine a strength of a match between the classification code and the text based on the match ratio and the proximity factor.
14. The computer readable medium of claim 13 , wherein the classification code includes an ICD-O-3 code and a behavior code.
15. The computer readable medium of claim 13 , comprising instructions causing the one or more processors to determine a behavior factor according to the following equation:
if the behavior code is “3” or if the behavior code is “2” and the text includes “non-invasive”, then the behavior factor is 1.5;
otherwise, the behavior factor is 1.0; and
wherein the instructions cause the one or more processors to multiply the match ratio, the proximity factor, and the behavior factor.
16. The computer readable medium of claim 13 , wherein determining the match ratio comprises determining the match ratio according to the following equation:
match ratio=k 2 /m, where k is the total number of the matched keywords and where m is the total number of the keywords.
17. The computer readable medium of claim 13 , wherein determining the proximity factor comprises determining the proximity factor according to the following equation:
proximity factor=1/(1+e/50), where e is a largest value between (s-m) and 0, where s is the shortest span of tokens within the note token stream containing all the matched keywords, and where m is a total number of the keywords in the classification code.
18. The computer readable medium of claim 13 , wherein generating the keyword map comprises:
read a secondary keyword and an ignored token;
determine a primary keyword as any of the keywords that are not a required keyword, the secondary keyword, or the ignored token; and
determine whether the primary keyword matches a second note token within the note token stream.
19. The computer readable medium of claim 13 , wherein generating the note token stream comprises:
reading a special token; and
converting the text to lowercase unless the text matches the special token.
20. The compute readable medium of claim 13 , wherein generating the note token stream comprises:
obtaining a descriptive synonym; and
replacing the note token with the descriptive synonym based on the note token matching the descriptive synonym.
21. The computer readable medium of claim 13 , wherein generating the note token stream comprises:
splitting the text into a plurality of sentences;
identifying a negative term from one of the plurality sentences; and
deleting a portion of the sentence of the plurality of the sentences from the negative term to an end of the sentence.
22. The computer readable medium of claim 13 , wherein the keyword span is the shortest span of tokens within the note token stream including the matched keywords.
23. The computer readable medium of claim 13 , wherein determining the strength includes multiplying the match ratio and the proximity factor.
24. The computer readable medium of claim 13 , wherein the instructions comprise instructions causing the one or more processors to tokenize the text of the note to create the note token stream.