Information processing using primary and secondary keyword groups

1. An information processing method executed by a computer, the method comprising:

acquiring a plurality of primary keyword and secondary keyword groups;

acquiring primary document information including more than one primary keyword, and secondary document information including more than one secondary keyword created by a first user;

classifying each of the plurality of secondary keywords by a plurality of topics based on a topic model, the topics being units for grouping a plurality of keywords having a threshold probability of appearing together in document information;

estimating whether each primary keyword in the plurality of groups is a related keyword related to any topic having a classified secondary keyword or a mixed keyword unrelated to any of the topics on the basis of primary topics having assigned primary keywords and secondary topics having assigned secondary keywords, the estimating comprising:

acquiring a primary topic of the primary keyword;

calculating the extent of the topic match being the proportion of secondary topics assigned to one or more secondary keywords that are the same as the primary topic;

acquiring the mixed keyword ratio or the ratio of primary keywords estimated to be mixed keywords among primary keywords included in primary document information from all users; and

calculating the mixed keyword probability of the primary keyword being a mixed keyword on the basis of the extent of the topic match and the mixed keyword ratio; and

assigning a primary keyword estimated to be a related keyword to a topic having a classified secondary keyword in the same group, and assigning a primary keyword estimated to be a mixed keyword to any of the topics for classification instead of a dedicated mixed keyword topic such that the primary keyword can later be assigned to a topic if determined to be a related keyword for a different user than the first user.

2. The method of claim 1 , further comprising:

determining a topic to assign a primary keyword estimated to be a related keyword on the basis of the proportion of secondary keywords classified by each topic in the same group; and

determining a topic to assign a primary keyword estimated to be a mixed keyword irrespective of the proportions.

3. The method of claim 1 , further comprising determining whether a primary keyword is a related keyword or a mixed keyword on the basis of the mixed keyword probability.

4. The method of claim 1 , further comprising calculating the likelihood Ψ kt of the t th primary keyword t, wherein t is an integer >1, appearing in the k th topic, wherein k is a predetermined integer >1, in all sets of primary document information from a user.

5. The method of claim 4 , further comprising generating the probability θ dk of the k th topic k in each set of secondary document information d, wherein d>1 but less than the total number of sets of secondary document information being generated.

6. The method of claim 5 , further comprising calculating a primary keyword generation probability P(t|d, D) of a primary keyword t being assigned to a single set of secondary document information d by totaling the θ dk Ψ kt of each topic k in the set of secondary document data d.