发明名称 |
SYSTEM AND METHOD FOR PREDICTING USER BEHAVIORS BASED ON PHRASE CONNECTIONS |
摘要 |
A method and system for generating alerts based on term taxonomies. The method includes analyzing textual content to detect a plurality of phrases in the textual content; identifying each of the detected phrases as any of: a sentiment phrase, and a non-sentiment phrase, wherein each sentiment phrase includes at least one word describing a sentiment; associating each identified non-sentiment phrase with at least one of the identified sentiment phrases to create at least one term taxonomy; analyzing the at least one term taxonomy to determine whether each identified non-sentiment phrase is associated with any of: a highly positive sentiment, and a highly negative sentiment; and generating an alert, when it is determined that one of the non-sentiment phrases is associated with any of: a highly positive sentiment, and a highly negative sentiment. |
申请公布号 |
US2016371280(A1) |
申请公布日期 |
2016.12.22 |
申请号 |
US201615250050 |
申请日期 |
2016.08.29 |
申请人 |
Taykey Ltd. |
发明人 |
AVNER Amit;DROR Omer |
分类号 |
G06F17/30 |
主分类号 |
G06F17/30 |
代理机构 |
|
代理人 |
|
主权项 |
1. A method for generating alerts based on term taxonomies, comprising:
analyzing textual content to detect a plurality of phrases in the textual content; identifying each of the detected phrases as any of: a sentiment phrase, and a non-sentiment phrase, wherein each sentiment phrase includes at least one word describing a sentiment; associating each identified non-sentiment phrase with at least one of the identified sentiment phrases to create at least one term taxonomy; analyzing the at least one term taxonomy to determine whether each identified non-sentiment phrase is associated with any of: a highly positive sentiment, and a highly negative sentiment; and generating an alert, when it is determined that one of the non-sentiment phrases is associated with any of: a highly positive sentiment, and a highly negative sentiment. |
地址 |
Herzliya Pituach IL |