Generic personalized recommendation system and multivariate automatic profiling method
Patent Information
- Authority / Receiving Office
- BR · BR
- Patent Type
- Applications
- Current Assignee / Owner
- GENIUS INST DE TECNOLOGIA
- Publication Date
- 2006-09-05
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing information and service selection tools are restricted to single electronic systems, passive, and require continuous user intervention and knowledge, exacerbated by digital convergence.
A multi-variable method of automatic profile definition that considers various user characteristics across multiple electronic systems, combined with a generic personalized recommendation system to provide adaptive and active recommendations.
Provides high-quality, adaptive recommendations with minimal user intervention by leveraging diverse user data sources, enhancing the variety and relevance of suggested information and services across different electronic platforms.
Abstract
Description
Generic system for personalized recommendation and multivariable method for automatic profile definition. FIELD OF THE INVENTION The present invention relates to a GENERIC PERSONALIZED RECOMMENDATION SYSTEM AND MULTIVARIABLE METHOD FOR AUTOMATIC PROFILE DEFINITION, to be used in any electronic information-based or service-providing systems. More specifically, the present invention consists of a method for automatically defining user profiles that is based not only on isolated aspects of user preferences, but on the set of various aspects of their preferences, in order to define a set of filters to be used in the automatic selection of information and services of interest to users, which represents their profile. Furthermore, the generic personalized recommendation system allows information and services of interest to users to be automatically recommended to them in different ways and according to their profile. FUNDAMENTALS OF THE INVENTION The increasing amount of information and services available in some electronic format demands the use of a series of increasingly complex tools for selecting information, as described in patents W02004030360-A1, W0200221335-A1, W0200145408-A1, W0200241178-A1 and services such as described in patents US6064980-A, JP318375-A, EP1045318A1, US2003103233-A1. A few years ago, the main sources of information were television and radio, which offer a limited number of free-to-air channels, selected manually according to viewers' preferences. Cable television represented a major revolution in this context, as the number of channels offered multiplied, providing users with a new range of information possibilities, but at the same time making the selection of this information more complicated. Therefore, EPGs (Electronic Program Guides) were created as tools to facilitate this information selection, as described in patents W02004030360-A1, W0200241178-A1, and EP1045318-A1. Technological progress in telecommunications has enabled the provision of a range of services remotely. Services that were previously only available in specific locations, requiring users to travel, are now available in their own homes. The growing number of services has necessitated the availability of tools such as telephone directories and, more recently, directory assistance services for users. With the advent of the internet and its rapid popularization, a new world of information and services became available to the public. Search engines, among which the most famous is GOOGLE, have become indispensable for obtaining the desired information or service. The user-defined keywords are used to find the results. These sites perform searches based on keywords defined by the users. The common shortcomings of all current information and service selection tools, despite their evolution, are that they are restricted, as they apply to a single electronic system, and / or passive, as they depend on the continuous action of the user and, in many cases, on the user's knowledge of the selection process. Therefore, aiming to address the causes of the aforementioned deficiencies, which are exacerbated by digital convergence—which is rapidly unifying telephony, the internet, and all other means of communication, access to information, and service provision—a new, unrestricted, and active GENERIC PERSONALIZED RECOMMENDATION SYSTEM AND MULTIVARIABLE METHOD FOR AUTOMATIC PROFILE DEFINITION was created. BRIEF DESCRIPTION OF THE FIGURE Figure 1 shows how the GENERIC PERSONALIZED RECOMMENDATION SYSTEM and MULTIVARIABLE METHOD for AUTOMATIC PROFILE DEFINITION work. BRIEF DESCRIPTION OF THE INVENTION The present invention provides, firstly, a multivariate method for automatic profile definition, taking into account, in a unified way, various characteristics of the user profile, such as preferences for television and radio programs, internet browsing preferences, use of telephone services, Places and events frequented, purchases made, and profiles of people with whom they maintain contact. The present invention provides, secondly, a generic personalized recommendation system that, based on filters defined by the user's profile, searches different Internet databases and recommends information and services of interest to the user, such as television and radio programs, events, promotions of consumer goods, places to visit, websites, etc., depending on the user's location and the electronic system in use. The major innovations of the multivariate automatic profiling method are the improved profile quality, since different sources of information regarding user preferences are used, and a very low need for user intervention, who only needs to identify the devices they use daily. The major innovations of the generic personalized recommendation system are that it adapts to the electronic system the user is using at any given moment and allows for a wide variety of recommendations. DETAILED DESCRIPTION OF THE INVENTION The operation of the GENERIC PERSONALIZED RECOMMENDATION SYSTEM and MULTIVARIABLE METHOD FOR AUTOMATIC PROFILE DEFINITION is presented in Figure 1. Figure 1 shows an example of the implementation of the generic personalized recommendation system. Based on digital television, digital radio, internet browser, cell phone with GPS (Global Positioning System) and credit card transactions, other combinations of electronic systems and services may also be used. The different electronic systems used in the system (1) to (5) are associated with one or multiple users, depending on their nature. Digital television (1) and digital radio (2) are typical examples of multi-user systems, while the mobile phone (4) and the credit card (5) are typical examples of single-user systems. The internet browser (3) can be associated with a single user through an email account or an authentication process based on login and password. The usage data from all electronic systems (1) to (5) to which a user is associated are sent to a user-specific database (6), residing on a server accessible via the internet, and classified as personal or group data. This data is processed by a profile management device (7), which accesses specific internet sites (10) and interprets the data, feeding it back into the database (6), and generating the user profile (8). This processing varies according to the type of data stored, such as: Data indicating that a user watched a specific digital television channel at a particular time is compared with that channel's programming available on the internet. to determine what type of content is present in the program being watched; Data indicating the user's location at a specific time of day (based on their mobile phone's GPS) can be used to identify that they were at a specific restaurant, or at a specific movie theater watching a specific film, or at a specific event, or at a store selling specific items; Data indicating a credit card transaction is interpreted as an expense at a specific type of establishment; A series of contacts made with a specific person at a specific time is interpreted as an affinity in some aspect of their profile; etc. The profile management device (7) therefore interprets the information, classified according to its nature, contained in the database (6), feeding it back into it and defining, based on artificial intelligence techniques, the user profile. (8) . 0 user profile is used by a search device (9), which automatically searches the internet (10) for information and services of interest to the user, and recommends them to the user through the recommendation device (11), which identifies which electronic system is being used by the user and provides the recommendation through an interface specifically designed for each electronic system. The system automatically adjusts to the frequency of the user's need for recommendations based on the Ill feedback on whether or not the recommendation is accepted by the user, which is stored in the database (6). The above description of the present invention is presented for illustrative and descriptive purposes only. Furthermore, the description is not intended to limit the invention to the form disclosed herein. Consequently, variations and modifications consistent with the above teachings, and the relevant skill or knowledge of the art, are within the scope of the present invention. Therefore, the embodiments described above are intended to better explain the known ways of practicing the invention and to allow those skilled in the art to use the invention in such, or other, embodiments and with various modifications necessary for the specific applications or uses of the present invention. It is intended that the present invention includes all modifications and variations thereof, within the scope described in the report and claims. attached slings.
Claims
CLAIMS 1. An automatic profiling method characterized by being multivariable and based, in a unified way, on various characteristics stored in a database that is fed by various electronic systems, which define the user's profile, used in a wide variety of commercial, industrial and business methods.
2. Automatic profile definition method, according to claim 1, characterized by using multiple variables defined by the set of electronic systems that feed the database.
3. Automatic profile definition method, according to claims 1 and 2, characterized by defining a single profile for each user, based on the multiple variables stored in the database.
4. Automatic profiling method, according to claims 1 to 3, characterized in that the various characteristics that define the user's profile are preferably derived from television and radio programs, internet browsing preferences, use of telephone services, places and events frequented, purchases made and profiles of people with whom they maintain contact.
5. Automatic profile definition method, according to claims 1 to 4, characterized by the database storing information derived from multi-user or single-user systems, for the definition of a single user's profile.
6. Automatic profile definition method, according to claim 5, characterized in that the multi-user system consists of an electronic system capable of being used by more than one user, without distinguishing between the different users, such as digital television and digital radio.
7. Automatic profiling method, according to claim 5, characterized in that the single-user system consists of an electronic system capable of distinguishing its different users, such as a mobile phone and a credit card.
8. Automatic profile definition method, according to claims 1 to 7, characterized by the user profile being ultimately defined based on the repetition pattern of various characteristics stored in the database by the various electronic systems.
9. Automatic profiling method, according to claims 1 to 8, characterized by the electronic systems preferably being digital television, digital radio, internet browser, mobile phone with or without location system and credit card transactions, as well as other combinations of devices and services capable of providing information to define the user's profile.
10. Automatic profiling method, according to claim 1, characterized in that the business method is preferably a personalized recommendation service.
11. Generic personalized recommendation system characterized by automatically adjusting to the frequency of the user's need for recommendations based on feedback regarding acceptance or rejection of the recommendation, adapting to the electronic system in use by the user in real time, and possessing the following devices: a) Different electronic systems associated with one or multiple users; b) A specific database for each user, which receives usage data from all electronic systems used by the user; c) A profile management device that accesses specific websites on the internet, interprets the data based on this information, feeds it back into the database, and generates the user's profile; d) A profile storage device capable of storing, for each user, the information that defines their profile, received from the profile management device; (e) A search device that automatically searches the internet for information and services of interest to the user, based on the profile stored in item "d"; f) A recommendation device that processes the information and services found by the search device and creates a recommendation Personalized recommendations, in addition to identifying which electronic system the user is using at any given time and providing this recommendation in real time, through an interface specifically designed for each electronic system.
12. Generic personalized recommendation system, according to claim 11, characterized by the differentiated electronic systems optionally being constituted by: television with digital reception, radio with digital reception, internet browser, credit card transactions, cell phone and any other portable electronic devices or equipment with or without a location system, in addition to other combinations of devices and services capable of providing information to define the user's profile.
13. Generic personalized recommendation system, according to claim 11, characterized by the database residing on a server accessible via the internet, through the profile management device.
14. Generic personalized recommendation system, according to claim 11, characterized in that the recommendation for item "f" can be provided by any form of human-machine interface and by a wide variety of electronic systems, either pre-configured or when requested by the user.