System and methods for predicting user behaviors based on phrase connections
A method and system for predicting user behaviors based on term taxonomies are provided. The system comprises generating phrases respective of user generated content, wherein each phrase is a sentiment phrase or a non-sentiment phrase, each sentiment phrase including at least one word describing a sentiment; identifying at least one connection between at least two of the generated phrases, wherein each connection is a direct connection or a hidden connection; generating at least one term taxonomy based on the identified at least one connection, wherein each term taxonomy is an association between a non-sentiment phrase and at least one of a plurality of sentiment phrases; periodically analyzing the at least one term taxonomy to determine at least one trend of each non-sentiment phrase respective of the associated plurality of sentiment phrases; and generating a prediction of future behavior of the at least one trend with respect to the at least one term taxonomy.
1. A method for predicting user behaviors based on term taxonomies, comprising:
generating phrases respective of user generated content, wherein each phrase is a sentiment phrase or a non-sentiment phrase, each sentiment phrase including at least one word describing a sentiment;
identifying at least one connection between at least two of the generated phrases, wherein each connection is a direct connection or a hidden connection;
generating at least one term taxonomy based on the identified at least one connection, wherein each term taxonomy is an association between a non-sentiment phrase and at least one of a plurality of sentiment phrases;
periodically analyzing the at least one term taxonomy to determine at least one trend of each non-sentiment phrase respective of the associated plurality of sentiment phrases; and
generating a prediction of future behavior of the at least one trend with respect to the at least one term taxonomy.
2. The method of claim 1 , wherein identifying at least one connection between at least two of the generated phrases further comprises:
identifying the directly connected phrases; and
filtering out the identified directly connected non-sentiment phrases to identify the hidden connections.
3. The method of claim 2 , wherein filtering out the identified directly connected non-sentiment phrases to identify the hidden connections further comprises:
analyzing the hidden connections to determine whether at least two hidden connections share a common non-sentiment phrase; and
upon determining that at least two analyzed hidden connections share a common non-sentiment phrase, identifying a new hidden connection respective of the at least two analyzed hidden connections.
4. The method of claim 1 , further comprising:
determining a context of the generated phrases.
5. The method of claim 4 , further comprising:
clustering a plurality of the generated phrases into at least one group based on the determined context.
6. The method of claim 1 , further comprising:
determining a probability that at least a first phrase and at least a second phrase of the generated phrases appear together.
7. The method of claim 1 , wherein each sentiment is any of: a positive sentiment, a neutral sentiment, and a negative sentiment.
8. The method of claim 1 , further comprising:
predicting an expected volume of the at least one trend at a given time;
tracking an actual volume of the at least one trend; and
computing a surprise factor; and
determining, based on the computed surprise factor, a general direction of the at least one trend using a grading function.
9. The method of claim 8 , wherein the grading function is a difference between the actual volume and the expected volume of the at least one trend.
10. A non-transitory computer-readable medium having stored thereon instructions for causing one or more processing units to execute the method according to claim 1 .
11. A system for predicting user behaviors based on term taxonomies, comprising:
a processing unit; and
a memory, the memory containing instructions that, when executed by the processing unit, configure the system to:
generate phrases respective of user generated content, wherein each phrase is a sentiment phrase or a non-sentiment phrase, each sentiment phrase including at least one word describing a sentiment;
identify at least one connection between at least two of the generated phrases, wherein each connection is a direct connection or a hidden connection;
generate at least one term taxonomy based on the identified at least one connection, wherein each term taxonomy is an association between a non-sentiment phrase and at least one of a plurality of sentiment phrases;
periodically analyze the at least one term taxonomy to determine at least one trend of each non-sentiment phrase respective of the associated plurality of sentiment phrases; and
generate a prediction of future behavior of the at least one trend with respect to the at least one term taxonomy.
12. The system of claim 11 , wherein the system is further configured to:
identify the directly connected phrases; and
filter out the identified directly connected non-sentiment phrases to identify the hidden connections.
13. The system of claim 12 , wherein the system is further configured to:
analyze the hidden connections to determine whether at least two hidden connections share a common non-sentiment phrase; and
upon determining that at least two analyzed hidden connections share a common non-sentiment phrase, identify a new hidden connection respective of the at least two analyzed hidden connections.
14. The system of claim 11 , wherein the system is further configured to:
determine a context of the generated phrases.
15. The system of claim 14 , wherein the system is further configured to:
clustering a plurality of the generated phrases into at least one group based on the determined context.
16. The system of claim 11 , wherein the system is further configured to:
determine a probability that at least a first phrase and at least a second phrase of the generated phrases appear together.
17. The system of claim 11 , wherein each sentiment is any of: a positive sentiment, a neutral sentiment, and a negative sentiment.
18. The system of claim 11 , wherein the system is further configured to:
predict an expected volume of the at least one trend at a given time;
track an actual volume of the at least one trend; and
compute a surprise factor; and
determine, based on the computed surprise factor, a general direction of the at least one trend using a grading function.
19. The system of claim 18 , wherein the grading function is a difference between the actual volume and the expected volume of the at least one trend.