Methods and apparatus to abstract events in software applications or services
View Patent ↗According to some embodiments, a system may be monitored to detect change events. A sequence associated with the detected change events may then be stored. The sequence may then be modified by deleting information associated with a detected change event. The sequence might also (or instead) be modified by adding information associated with a non-detected change event. Information associated with the normalized sequence may then be provided.
1. A computer-implemented method, comprising:
monitoring a system to detect change events, each change event indicating an interaction between a user and data objects within an application;
defining a set of abstract action types, wherein each abstract action type is associated with a type of interaction between a user and an application;
mapping a detected change event into an appropriate abstract action type;
associating a plurality of actions with each detected change event;
storing an ordered sequence of the actions associated with the detected change events, wherein the sequence comprises an ordered sequence of interactions between the user and the application associated with the system;
normalizing the sequence, wherein normalizing comprises:
determining if a first action of the sequence is identical to a second action of the sequence;
deleting the first action from the sequence if the first action is identical to a second action;
providing information associated with the normalized sequence to an analyzer wherein the provided information is analyzed to determine one or more pieces of relevance evidence describing a relationship between data objects;
determining the one or more pieces of relevance evidence between data objects based on the normalized sequence, wherein determining the one or more pieces of relevance evidence comprises:
retrieving the stored sequence of actions from a data store;
determining one or more pieces of relevance evidence by identifying patterns associated with the interaction between the user and the application based on the retrieved actions;
assigning a rank to each piece of relevance evidence; and
sorting the one or more pieces of relevance evidence based on the assigned rank,
wherein each piece of relevance evidence comprises (i) a type to identify each piece of relevance evidence, (ii) a weight to identify an importance of each piece of relevance evidence, and (iii) a confidence to identify a reliability of each piece of relevance evidence;
determining a first data object of interest;
based on the relevance evidence, selecting a second data object associated with the first data object; and
presenting to the user a list of related data objects, including the second data object.
2. The method of claim 1 , wherein a change event is associated with at least one of: (i) a user interaction with an application, (ii) a user interaction with a data object, (iii) an action automatically generated by an application, (iv) a service, (v) a server, or (vi) a peripheral.
3. The method of claim 1 , wherein a detected change event is discarded if it is not associated with an abstract action type.
4. The method of claim 1 , wherein the set of abstract action types includes at least one of: (i) open data object; (ii) close data object; (iii) activate data object; (iv) deactivate data object; (v) create data object; (vi) delete data object; (vii) copy data object; (viii) replace data object; (ix) active hyperlink in data object; (x) copy content from data object; (xi) paste content within data object; and (xii) notification that data object is ready.
5. The method of claim 1 , wherein said normalizing includes:
ensuring that an open data object action is followed by an activate data object action of the same data object.
6. The method of claim 1 , wherein said normalizing includes:
ensuring that an activate data object action is preceded by a deactivate data object action of a previously active data object.
7. The method of claim 1 , wherein said normalizing includes at least one of:
(i) ensuring that a deactivate data object action is ignored if the data object is not active, (ii) ensuring that an activate data object action is ignored if the data object is not open, (iii) ensuring that a deactivate data object action is ignored if the data object is not open, and (iv) ensuring that a close data object action is ignored if the data object is not open.
8. The method of claim 1 , further comprising:
archiving the sequence in an order based on at least one relevant parameter.
9. The method of claim 8 , wherein the relevant parameter is associated with a chronological order in which the change events occurred.
10. The method of claim 1 , wherein the sequence includes at least one of: (i) an action type, (ii) a data source, (iii) a time stamp, or (iv) an action name.
11. The method of claim 1 , wherein the first data object is of interest to the user.
12. A computer memory storing instructions adapted to be executed by a processor to perform a method, said method comprising:
monitoring a system to detect change events, each change event indicating an interaction between a user and data objects within an application;
defining a set of abstract action types, wherein each abstract action type is associated with a type of interaction between a user and an application;
mapping a detected change event into an appropriate abstract action type;
associating a plurality of actions with each detected change event;
storing an ordered sequence of the actions associated with the detected change events, wherein the sequence comprises an ordered sequence of interactions between the user and the application associated with the system;
normalizing the sequence, wherein normalizing comprises:
determining if a first action of the sequence is identical to a second action of the sequence;
deleting the first action from the sequence if the first action is identical to a second action;
providing information associated with the normalized sequence to an analyzer wherein the provided information is analyzed to determine one or more pieces of relevance evidence describing a relationship between data objects;
determining the one or more pieces of relevance evidence between data objects based on the normalized sequence, wherein determining the one or more pieces of relevance evidence comprises:
retrieving the stored sequence of actions from a data store;
determining one or more pieces of relevance evidence by identifying patterns associated with the interaction between the user and the application based on the retrieved actions;
assigning a rank to each piece of relevance evidence; and
sorting the one or more pieces of relevance evidence based on the assigned rank,
wherein each piece of relevance evidence comprises (i) a type to identify each piece of relevance evidence, (ii) a weight to identify an importance of each piece of relevance evidence, and (iii) a confidence to identify a reliability of each piece of relevance evidence;
determining a first data object of interest;
based on the relevance evidence, selecting a second data object associated with the first data object; and
presenting to the user a list of related data objects, including the second data object.
13. The medium of claim 12 , a change event is associated with at least one of: (i) a user interaction with an application, (ii) an action automatically generated by an application, or (iii) a service.
14. An apparatus, comprising:
a monitoring component to detect change events in a system, each change event indicating an interaction between a user and data objects within an application, to define a set of abstract action types, wherein each abstract action type is associated with a type of interaction between a user and an application, to mare a detected change event into an appropriate abstract action type, to associate a plurality of actions with each detected change event, and to store a sequence of the actions associated with the detected change events, wherein the sequence comprises a sequence of interactions between the user and the application associated with the system; and
a normalizing component to normalize the sequence associated with detected change events, wherein normalizing comprises:
determining if a first action of the sequence is identical to a second action of the sequence;
deleting the first action from the sequence if the first action is identical to a second action;
wherein the normalizing component is further to provide information associated with the normalized sequence to an analyzer, wherein the provided information is analyzed to determine relevance evidence between data objects, and based on the relevance evidence a first data object of interest is to be determined and a second data object associated with the first data object is to be selected, wherein a list of related data objects, including the second data object is to be presented to the user, wherein relevance evidence is to be determined between data objects based on the normalized sequence,
wherein determining the relevance evidence comprises:
retrieving the stored sequence of actions from a data store;
determining one or more pieces of relevance evidence by identifying patterns associated with the interaction between the user and the application based on the retrieved actions;
assigning a rank to each piece of relevance evidence; and
sorting the one or more pieces of relevance evidence based on the assigned rank,
wherein each piece of relevance evidence comprises (i) a type to identify each piece of relevance evidence, (ii) a weight to identify an importance of each piece of relevance evidence, and (iii) a confidence to identify a reliability of each piece of relevance evidence.
15. The apparatus of claim 14 , further including:
an evidence store to store the evidence or relationships between data objects,
wherein the analyzer is to receive from the normalizing component information associated with the normalized sequence and is to determine evidence of relationships between data objects.