Document analysis system, document analysis method, and document analysis program

1. A computer implemented e-mail analysis system comprising:

a keyword database that retains a plurality of keywords, each of the keywords having a correspondence with one or more classification codes, the classification codes being used to classify one or more e-mails;

a processor; and

a memory, wherein the memory has stored thereon instructions that, when executed by the processor, causes the processor to:

search one of the keywords in a first e-mail and assign a first classification code corresponding to the keyword;

accept a second classification code assigned by a user for a second e-mail to which no classification code has been assigned;

calculate a score, which indicates strength of linkage between the first e-mail and the first classification code as well as between the second e-mail and the second classification code, and quantifies priority of the e-mails to be monitored, on the basis of a plurality of terms that respectively appear in the first e-mail and the second e-mail to which the one or more classification codes have been assigned and an evaluation value of each of the plurality of terms;

learn the evaluation value of each of the plurality of terms based on a process of classifying the one or more e-mails, the learned evaluation value being used to calculate the score;

prepare an investigation report by using the calculated score when the calculated score exceeds a specified threshold value arbitrarily set by the system on the basis of an input from the user; and

promote efficiency in monitoring the e-mails by feeding back the score to the user by outputting the prepared investigation report.

2. The e-mail analysis system according to claim 1 ,

wherein the memory has stored thereon further instructions that, when executed by the processor, causes the processor to:

extract the e-mail having the score which is fed back to the user;

the system further comprising:

an e-mail display unit that displays the extracted e-mail on a document display screen.

3. The e-mail analysis system according to claim 1 ,

wherein the second classification code is assigned by the user to the e-mail on the basis of a relation with a specified case.

4. The e-mail analysis system according to claim 3 ,

wherein the memory has stored thereon further instructions that, when executed by the processor, causes the processor to:

accept input of a category of the specified case.

5. The e-mail analysis system according to claim 1 ,

wherein the memory has stored thereon further instructions that, when executed by the processor, causes the processor to:

further calculate the score in a time-series manner; and

calculate a moving average of a group of scores calculated in the time-series manner as the score indicative of the strength of linkage between the one or more e-mails and the first and the second classification codes.

6. An e-mail analysis method executed by a computer, comprising:

referring to a keyword database retaining a plurality of keywords, each of the keywords having a correspondence with one or more classification codes, the classification codes being used to classify one or more e-mails;

searching one of the keywords in a first e-mail and assign a first classification code corresponding to the keyword;

accepting a second classification code assigned by a user for a second e-mail to which no classification code has been assigned;

calculating a score, which indicates strength of linkage between the first e-mail and the first classification code as well as between the second e-mail and the second classification code, and quantifies priority of the e-mails to be monitored, on the basis of a plurality of terms that respectively appear in the first e-mail and the second e-mail to which the one or more classification codes have been assigned and an evaluation value of each of the plurality of terms;

learning the evaluation value of each of the plurality of terms based on a process of classifying the one or more e-mails, the learned evaluation value being used to calculate the score;

preparing an investigation report by using the calculated score when the calculated score exceeds a specified threshold value set on the basis of an input from the user; and

promoting efficiency in monitoring the e-mails by feeding back the score to the user by outputting the prepared investigation report.

7. A non-transitory computer readable storage medium having stored thereon an e-mail analysis program for causing a computer to:

refer to a keyword database retaining a plurality of keywords having a correspondence with one or more classification codes, the classification codes being used to classify one or more e-mails;

search one of the keywords in a first e-mail and assign a first classification code corresponding to the keyword;

accept a second classification code assigned by a user for a second e-mail to which no classification code has been assigned;

calculate a score, which indicates strength of linkage between the first e-mail and the first classification code as well as between the second e-mail and the second classification code, and quantifies priority of the e-mails to be monitored, on the basis of a plurality of terms that respectively appear in the first e-mail and the second e-mail to which the one or more classification codes have been assigned and an evaluation value of each of the plurality of terms;

prepare an investigation report by using the calculated score when the calculated score exceeds a specified threshold value arbitrarily set on the basis of an input from the user; and

promote efficiency in monitoring the e-mails by feeding back the score to the user by outputting the prepared investigation report.