Sentiment normalization using personality characteristics

A system comprising:

• a computer readable storage medium storing program instructions;
• and
• a processor configured to execute the program instructions to perform a method comprising:
• characterizing personality profiles of a plurality of authors, wherein characterizing the personality profiles further comprises:
• computing scores of a plurality of metrics comprising agreeableness, extraversion, conscientiousness, openness, neuroticism, helpfulness, stimulation, tradition, pleasure, achievement, harmony, love, practicality, liberty, and closeness for each of the plurality of authors;
• and
• determining a percentile of each score of each metric for each of the plurality of authors relative to a population having a respective distribution of scores for each of the plurality of metrics;
• clustering a set of authors into a first cluster based on the personality profiles of the plurality of authors, wherein the first cluster is defined by a respective range of scores for each of the plurality of metrics;
• compiling, in response to the clustering, a first set of textual data created by authors of the set of authors in the first cluster;
• computing a plurality of sentiment scores of a plurality of portions of text from the first set of textual data;
• normalizing the plurality of sentiment scores, wherein normalizing the plurality of sentiment scores further comprises:
• computing a sum of sentiment scores for the first cluster by adding the plurality of sentiment scores generated for the first cluster;
• computing an average sentiment score by dividing the sum of sentiment scores for the first cluster by a total number of sentiment scores generated for the first cluster;
• and
• computing a plurality of normalized sentiment scores by dividing each of the plurality of sentiment scores for the first cluster by the average sentiment score of the first cluster;
• and
• storing the plurality of normalized sentiment scores for the first cluster and the respective range of scores for each of the plurality of metrics defined by the first cluster in the computer readable storage medium.

A computer program product comprising a computer readable storage medium having program instructions embodied therewith, wherein the computer readable storage medium is not a transitory signal per se, the program instructions executable by a processor to cause the processor to perform a method comprising:

• characterizing personality profiles of a plurality of authors, wherein characterizing the personality profiles further comprises:
• computing scores of a plurality of metrics comprising agreeableness, extraversion, conscientiousness, openness, neuroticism, helpfulness, stimulation, tradition, pleasure, achievement, harmony, love, practicality, liberty, and closeness for each of the plurality of authors;
• and
• determining a percentile of each score of each metric for each of the plurality of authors relative to a population having a respective distribution of scores for each of the plurality of metrics;
• clustering a set of authors into a first cluster based on the personality profiles of the plurality of authors, wherein the first cluster is defined by a respective range of scores for each of the plurality of metrics;
• compiling, in response to the clustering, a first set of textual data created by authors of the set of authors in the first cluster;
• computing a plurality of sentiment scores of a plurality of portions of text from the first set of textual data;
• normalizing the plurality of sentiment scores, wherein normalizing the plurality of sentiment scores further comprises:
• computing a sum of sentiment scores for the first cluster by adding the plurality of sentiment scores generated for the first cluster;
• computing an average sentiment score by dividing the sum of sentiment scores for the first cluster by a total number of sentiment scores generated for the first cluster;
• and
• computing a plurality of normalized sentiment scores by dividing each of the plurality of sentiment scores for the first cluster by the average sentiment score of the first cluster;
• and
• storing the plurality of normalized sentiment scores for the first cluster and the respective range of scores for each of the plurality of metrics defined by the first cluster in the computer readable storage medium.