| 摘要 |
A decision-theoretic approach for guiding automated actions. In one embodiment, a computer-implemented method first determines a text to analyze. The method then determines an action probability based on the text. Words and phrases as well as information in special fields, such as distinctions and patterns in the header of email, observations about recent user activity, and organizational information are considered in classification. Based on the inferred probability that a user has a goal under consideration, the method selects one of the following options: (1) inaction, (2) automatic action, or (3) suggested action with user approval. Upon the method selecting either the (1) automatic action option or the (2) suggested action with user approval option-the latter also in conjunction with receiving actual user approval-the method performs an action based on the text.
|
