Systems and methods for directing automated services for messaging and scheduling
First Claim
1. A computer-implemented method comprising:
- determining a message to analyze;
determining a probability that the message should be scheduled based at least in part on the message and contextual information;
providing a selection option of at least one of;
(1) inaction, (2) automatic action, and (3) suggested action with user approval; and
performing a scheduling action based at least in part on the scheduling probability and a selected option, wherein selecting an option comprises;
associating a boundary between the (1) and the (2) options with a first probability threshold, and a boundary between the (2) and the (3) options with a second probability threshold;
upon determining that the scheduling probability is less than the first probability threshold, selecting the (1) option;
upon determining that the scheduling probability is greater than the second probability threshold, selecting the (3) option; and
, othewise, selecting the (2) option.
2 Assignments
0 Petitions
Accused Products
Abstract
Directing automated services for messaging and scheduling. In one embodiment, a computer-implemented method first determines a message to analyze. The method then determines the probability that the user would like to review a calendar or to schedule an appointment based on information in the message, based on the message, and based on the scheduling probability, 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 a scheduling action based on the message in a manner that takes into consideration the precision and confidence of the scheduling action.
-
Citations
32 Claims
-
1. A computer-implemented method comprising:
-
determining a message to analyze;
determining a probability that the message should be scheduled based at least in part on the message and contextual information;
providing a selection option of at least one of;
(1) inaction, (2) automatic action, and (3) suggested action with user approval; and
performing a scheduling action based at least in part on the scheduling probability and a selected option, wherein selecting an option comprises;
associating a boundary between the (1) and the (2) options with a first probability threshold, and a boundary between the (2) and the (3) options with a second probability threshold;
upon determining that the scheduling probability is less than the first probability threshold, selecting the (1) option;
upon determining that the scheduling probability is greater than the second probability threshold, selecting the (3) option; and
, othewise, selecting the (2) option.- View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16)
performing a feature-selection phase; and
performing a model-construction phase based on the feature-selection phase.
-
-
10. The method of claim 9, wherein performing a feature-selecting phase comprises training the text classification system by inputting a plurality of messages based on which scheduling is to occur and inputting a plurality of messages based on which scheduling is not to occur.
-
11. The method of claim 9, wherein performing a feature-selection phase comprises seeding the feature-selection phase with domain-specific knowledge.
-
12. The method of claim 11, wherein the domain-specific knowledge comprises at least one of:
- multiword phrases, nature of a relationship between a sender and a receiver, a number of recipient of the message, and Boolean combinations of variables composed by joining multiword phrases with at least one of a date variable and a tie variable.
-
13. The method of claim 1, wherein associating a boundary between the (1) and the (2) options with a first probability threshold, and a boundary between the (2) and the (3) options with a second probability threshold comprises utilizing decision theory to determine the first and the second probability thresholds.
-
14. The method of claim 13, wherein utilizing decision theory comprises utilizing cost/benefit analysis.
-
15. The method of claim 14, wherein utilizing cost/benefit analysis comprises utilizing the cost/benefit analysis in a contextual manner.
-
16. The method of claim 1, wherein performing a scheduling action comprises:
-
determining an anchor date of the message; and
,parsing a text of the message relative to the anchor date against typical commonsense patterns and assumptions of typical language usage.
-
-
17. A machine-readable medium having processor instructions stored thereon for execution by a processor to cause performance of a method comprising:
-
determining a message to analyze;
determining a probability that the message should be scheduled;
providing a selection option of at least one of;
(1) inaction, (2) automatic action, and (3) suggested action with user approval; and
performing a scheduling action based at least in part on the scheduling probability and a selected option, wherein selecting an option comprises;
associating a boundary between the (1) and the (2) options with a first probability threshold, and a boundary between the (2) and the (3) options with a second probability threshold;
upon determining that the scheduling probability is less than the first probability threshold, selecting the (1) option;
upon determining that the scheduling probability is greater than the second probability threshold, selecting the (3) option; and
,otherwise, selecting the (2) option. - View Dependent Claims (18, 19, 20, 21, 22, 23, 24, 25, 26, 27)
performing a feature-selection phase; and
,performing a model-construction phase based on the feature-selection phase.
-
-
22. The medium of claim 21, wherein performing a feature-selecting phase comprises training the text classification system by inputting a plurality of messages based on which scheduling is to occur and inputting a plurality of messages based on which scheduling is not to occur.
-
23. The medium of claim 21, wherein performing a feature-selection phase comprises seeding the feature-selection phase with domain-specific knowledge.
-
24. The medium of claim 17, wherein associating a boundary between the (1) and the (2) options with a first probability threshold, and a boundary between the (2) and the (3) options with a second probability threshold comprises utilizing decision theory to determine the first and the second probability thresholds.
-
25. The medium of claim 24, wherein utilizing decision theory composes utilizing cost/benefit analysis.
-
26. The medium of claim 25, wherein utilizing cost/benefit analysis comprises utilizing the cost/benefit analysis in a contextual manner.
-
27. The medium of claim 25, wherein performing a scheduling action comprises:
-
determining an anchor date of the message; and
,parsing a text of the message relative to the anchor date against typical patterns and assumptions of common-sense language.
-
-
28. A scheduling system, comprising:
-
a message with associated content;
an analyzing component that determines a probability that the message should be scheduled based on the message content; and
a scheduling component that configures a schedule based on the scheduling probability. - View Dependent Claims (30, 31, 32)
-
-
29. A scheduling system, comprising:
-
a message with associated content;
an analyzing component that determines a probability that the message should be scheduled based on the message content; and
a scheduling component that configures a schedule based On the scheduling probability, the analyzing component employing the following equation in determining the scheduling probability;
-
Specification