This patent us7930197 says data mining of personal data is patented. Looks like patent officers weren't aware of data mining's huge overreach.
This patent claims:
A personal user information data mining system, comprising the following computer-implemented components:
a processor; a memory communicatively coupled to the processor; a data repository adapted to store personal user data supplied by a plurality of human users through a plurality of user devices communicatively coupled to the personal user information data mining system over a network, the personal user data comprising information generated by the human users, information owned by the human users, and information linked to the human users;
a data configuration component that executes in the processor from the memory and, when executed by the processor, normalizes the personal user data from a plurality of disparate taxonomies into a single taxonomy;
a data mining component that executes in the processor from the memory and, when executed by the processor, identifies at least one correlation from the normalized personal user data; and
an application component that executes in the processor from the memory and, when executed by the processor, (a) retrieves the personal user data from the plurality of user devices, (b) stores the retrieved personal user data in the data repository, (c) retrieves the identified correlation, and (d) provides the plurality of human users with the identified correlation through the plurality of user devices, (e) determines a user action based on the said correlation, (f) determines a value of the user action based on said correlation, (g) and notifies a user of the user action when the value of the user action is greater than the cost of interruption.
Which is not the same as claiming:
Datamining of personal data.
Firstly, remember that each and every element of the claim must be adopted to infringe a claim.
There is a main "data repository" containing all users data.
In layman's terms, the most important distinguishing aspects of the invention are steps (d)-(g).
Personal data on a device is compared with a main data repository ("identifies at least one correlation from the normalized personal user data"), e.g. user bob's current project & bob's location is identified as having a correlation with user jill's project and location.
(e): "determines a user action based on the said correlation" : e.g. send an SMS to Bob, to helpfully suggest to him that he might want to meet Jill while he is in Texas.
(f) lets say this is of 8/10 value
(g) notifies a user of the user action when the value of the user action is greater than the cost of interruption: lets say cost of interruption is 7, we above the threshold with the project/location Jill correlation; so the system decides to send the SMS to Bob.
The system would raise many privacy issues. It essentially seeks to identify the correlation from personal data with other data, value it, then take a user action automatically if it is above a threshold. e.g. you might use an receipt for an outdoor concert in NY Central Park, to then notify the user that there is a thunderstorm forecast on the day they are attending an outdoor concert; or so forth.