User profiling is gaining in importance in an increasingly digital world.

An example of a company that depends on building of user profiles to establish relationships with customers is the on-line book retailer Amazon. com, Inc.

Internet sites like Amazon try to build a profile of a users interest on. the basis of their browsing and purchasing behaviour. This is a relatively poor and incomplete measure of a user's interest because buying a certain book does not necessarily directly reflect a user's interest. Furthermore, profile building can only be done on the basis of the observed user interaction behaviour on a single and specific site, which hinders the formation of accurate user models. Similar difficulties are experienced by recommender systems for consumer electronics such as video and audio electronic programme guides. Profiling could be improved by allowing services and applications to share data on a specific user. Whilst technically feasible, such sharing raises privacy and data protection issues and is unlikely to be popular with users.

It is an object of the present invention to enable the improved generation of user profiles whilst at the same time addressing privacy issues.

In accordance with the first aspect of the present invention there is provided an interactive system comprising a plurality of devices with which a user may interact, a personal device of the user configured to log data about the users interaction, and at least one further device with which the user may interact, said at least one further device adapting the interaction on the basis

of user interaction data held by said personal device. By use of a personal device to provide a repository for the interaction data from a number of interactions, a broader range of data is available to permit more accurate adaption and/or user profiling. Furthermore, with the personal device, the user maintains ownership of, and control over, their own interaction data.

Each of the said plurality of devices may be configured to provide user interaction data to said personal device according to a predetermined data format, for example XML.

The invention further provides a portable device for use as said personal device in the system recited above, the portable device comprising data input, processing, storage and output means configured to log data relating to user interaction with a plurality of other devices and to make said data available to a further device. The portable device may, further comprise user operable input means coupled with the data processing means by operation of which a user is enabled to control the making available of stored interaction data to said further device and/or enabled to modify stored interaction data prior to supply to said further device.

In the portable device each of said data input and data output means is suitably configured to handle data transmission via a predetermined transmission mechanism, such as a wireless radio link operating according to a predetermined protocol (for example Bluetooth or ZigBee).

The portable device may be constructed integrally with one of said plurality of devices, said one being itself portable: for example, the device may be built into a mobile phone or PDA. Alternatively, the input and output means may respectively be mechanically configured to cooperate with output means of said plurality of devices and input means of said at least one further device, such as to allow the portable device to be plugged into (or otherwise cooperate with) the device being interacted with. In such an arrangement, the portable device may be configured as a smart card.

Also in accordance with the present invention there is provided a method of managing user interactions in a system comprising a plurality of devices with which a user may interact, wherein a personal device of the user

logs data about the users interaction, and in interactions with at least one further device, said at least one further device adapts the interaction on the basis of user interaction data held by said personal device.

In such a method, the said at least one further device may apply a predetermined set of rules to create a user profile based on the user interaction data and, where the said at least one further device stores the created user profile, on a subsequent interaction with that user, it may recall the profile and update it based on the interaction data held by the personal device.

The invention further provides a computer program product directly loadable into the processor means of a portable device according to the invention, comprising software code portions for causing the processor means to format received interaction data, store it in predetermined format in the storage means and send it to the output means on request from an at least one further device.

Further aspects of the present invention are described in the following description of preferred embodiments of the invention.

The present invention will now be described with reference to the figures of the accompanying drawings in which: Figure 1 schematically represents a personal device embodying the present invention performing a series of interactions with a number of further devices; Figure 2 represents the organisation of data in a database held in the personal device of Figure 1; Figure 3 shows functional components of a personal device suitable to embody the present invention; and Figure 4 represents an alternative embodiment of the personal device of Figure 1 in the form of a smart card for use with a mobile telephone.

It should be noted that the drawings are diagrammatic and not drawn to scale. Relative dimensions and proportions of parts of the figures have been

shown exaggerated or reduced in size for the sake of clarity and convenience in the drawings. The same reference signs are generally used to refer to corresponding or similar features in a different embodiment.

Referring to Figure 1, a series of interactions (identified by arrows 1 to 3) are represented. The user is represented by a personal portable device 10 which logs data relating to each interaction in an internally held store or database 12. The devices 14,16, 18 with which the user interacts are suitably configured to deliver to the portable device 10 the interaction data in a predetermined format: various mechanisms for the delivery of data to the portable device will be discussed below.

In a typical scenario, the first device 14 with which the user interacts may be a supermarket checkout, downloading to the portable device 10 an identifier for the supermarket, a list of the goods purchased and possibly further data. The second device 16 with which the user interacts may be an automated ticketing facility for a cinema, downloading to the portable device 10 an identifier for the cinema and the film requested, again possibly with further data.

The third device 18 with which the user interacts may be a terminal for an on-line shopping service (perhaps a users home personal computer, or a dedicated device). At the start of the interaction, the device 18 determines the presence of the personal device 10, sends a request to it (suitably using the same mechanism as used to transfer user interaction data to the portable device) and uploads the user interaction data from database 12. On receipt, the device applies a predetermined set of rules to the raw interaction data to create a profile for the user. Having created the profile, the device selects a series of product recommendations based on that profile and presents them to the user. By basing the profile on data from more than one source (devices 14,16) greater accuracy in product recommendation should result, as well as a greater range of recommendations likely to be of interest. As the interaction with device 18 proceeds, further interaction data is created and again logged in database 12.

As will be appreciated, the first and second devices 14,16 may also create profiles and adapt the interactions accordingly based on previously acquired data stored in database 12. Also, rather than creating a full profile and generating recommendations accordingly, the adaption may be at a simpler level, for example not offering to a user items that they have already acquired (and are not likely to require again).

Figure 2 represents a suitable scheme for storage of the interaction data in the database 12 (Fig. 1). At a first indexing level 20, interaction data is sorted and separated from any other information that may be stored in the database. At the next level the information is sorted by category 22,24 : in the example scenario of Figure 1, Category 1 (22) is shopping, with data from the interactions with devices 14 and 18 stored under this heading. Category 2 (24) is cinema with data from the interaction with device 16 stored.

Below each Category is a first layer of Sub-Category 26,28, 30. Under the shopping Category 22, the first Sub-Category may comprise a division into over-the-counter shopping 26 (the interaction with device 14) and online shopping (via device 18). Under each first Sub-Category are a number of second Sub-Categories 32,34, 36,38. For the first Sub-Category 26 (over- the-counter shopping) the second Sub-Category 32,34 may identify individual stores, or individual visits to stores (or a further layer of Sub-Categorisation may be provided to allow both to be stored separately). Under each of the second Sub-Categories 32,34, 36,38 is stored the respective logged interaction data 40,42, 44,46.

Under the Category 24 (cinema) the first Sub-Category 30 may be the identity of the particular cinema, and the second Sub-Category 38 the date of the visit, with the data 46 identifying the particular film viewed (or requested to be viewed). As will be recognised, other arrangements of database, perhaps including more or fewer levels of Sub-Categorisation in the arrangement described, may be used. The important criteria is that the data is stored in a regularised manner that makes it simple to store and to access. As will also be recognised, other forms and classes of interaction may be logged, specifically

those which will contribute interaction data on the basis of which user profiles may be drawn up.

Figure 3 schematically represents the functional components of the users personal device 10 (Fig. 1) with input 50 and output 52 means coupled with control means 54. A storage device 56 holds the database 12 (Fig. 1) created under the direction of the control means 54, and user input means 58 allow user control of operation of the personal device. Optionally, a display 60 or other output device may also be provided to allow the user to review stored interaction data or other data (for example status information) of the personal device.

The form of the input 50 and output 52 means will be determined by the mechanism used for transfer of data to and from the devices interacted with.

In a wireless embodiment, where the data is transferred by radio frequency or infra-red link, the input and output means comprise suitably configured transmitters/receivers (or may be combined in a transceiver). There are a number of standardised data transmission protocols that may be used, for example Bluetooth, ZigBee, IEE 802.11, Ultra Wideband (UWB), IrDA and, if a single form of transfer mechanism is not standardised or selected by the majority of users, multiple forms and configurations of input/output means may be provided in tandem (although this duplication will tend to increase the per- unit cost of the personal device). The arrangement of data for transfer should also be in standardised language: XML is suggested as a suitable candidate.

The control means 54 may suitably be provided by a programmable processor device: the processor power need not be particularly high as it is simply required to handle read/write duties from/to the storage means 56 for data in a predetermined format.

The storage means 56 comprises a non-volatile random access memory device, which may be solid state integrated memory device, or an optical or magnetic disk, although the mechanics of a disk read/write stage may increase the size/weight of the personal device to unacceptable levels.

The form and function of the user input device 58 will be determined by the control functions to be provided to the user. Where the user is to be simply

offered a"yes/no"decision on whether or not the personal device is to be involved in an interaction, the controls may be relatively simple. Such a choice may be appropriate if the user is about to take part in a non-representative interaction (such as making purchases on behalf of a third party) where the interaction data would dilute rather than contribute to the accuracy of profiling that may be achieved based on the stored data.

Where more complex functionality is to be supported, up to and including allowing the user to review and edit or augment the stored items of data, more complex controls up to a full alphanumeric input keyboard, XY pointer device or the like may be provided. The form of display device 60 will also depend on the level of control functions supported, from a simple indicator lamp or LED for"yes/no"indications, to a fuller liquid crystal or similar display screen if the user is to have the opportunity to review and amend stored interaction data.

The form and function of the component features shown in Figure 3 will also be determined by whether the personal device is a stand-alone device (with no function other than the logging of interaction data) or whether it is built into another portable device such as a mobile phone or personal digital assistant (PDA). In the latter case, where a device with which the user interacts is also the host for the personal device, some or all of the components may have roles for both devices (for example the store 56 may comprise a partitioned area of general memory for the device).

A further alternative arrangement is illustrated in Figure 4. Here the personal device 10 is in the form of a smart card which, in order to log interactions and permit access to the stored data, must be put into a suitable card reader in the device to be interacted with-mobile phone 62 in the example shown. Both mechanical and wireless means for connection to smart cards are well known in the art and will not be discussed in detail here.

In the foregoing we have presented a method for managing user interactions in a system comprising a plurality of devices with which a user may interact, together with devices configured to support the same. A personal device of the user logs data about the users interaction, and in

interactions with at least one further device, the further device adapts or personalises the interaction on the basis of user interaction data held by the personal device. The personal device is portable, or built into a portable device such as a mobile phone, and allows the user to maintain ownership of, and control over, their interaction data.

From reading the present disclosure, other modifications will be apparent to persons skilled in the art. Such modifications may involve other features which are already now in the design manufacture and use of data storage and processing devices and systems, and which may be used instead of or in addition to features already described herein.