SYSTEM AND METHOD FOR RECOMMENDATION SYSTEM BASED ON IMPLICIT FEEDBACK - Patent application

Abstract

A system and method are described for providing a recommender system for a vehicle content delivery network. A static recommendation list can be generated based on trip characteristics stored on a server. An efficiency threshold can be calculated or provided that sets the point at which the static recommendations are no longer used for recommendations and the recommender system can be used instead. Once data is collected about each user's preferences based on implicit feedback, the data can be analyzed to calculate an efficiency level for the recommender system operating on the data collected for a user. When the efficiency level meets or exceeds the efficiency threshold, the recommender system is automatically used for the particular user.

Description

[Technical field]

[0001] The field of the invention is recommendation systems and methods. [Background technology]

[0002] The following description includes information that may be useful in understanding the present invention. This is not an admission that any of the information provided herein is prior art or relevant to the presently claimed invention, or that any publication specifically or implicitly referenced is prior art.

[0003] Online stores and user-generated media platforms utilize recommendation systems to facilitate browsing through a large number of items. Such systems typically rely on user preferences and centrally stored data about the intrinsic properties of the items (i.e., content-based recommendations) or the preferences of other users (i.e., recommendations using collaborative filtering).

[0004] In some situations, it is best to combine the two systems above and dynamically choose the best approach depending on the situation (i.e., a hybrid recommendation). The efficiency of a recommendation system is generally limited by the data the system has access to about user preferences. When sufficient data is not available, a recommendation system often cannot operate efficiently. This situation is often referred to as the cold start problem.

[0005] There are two common approaches to collecting the required data: the first involves having new users fill out a survey and requesting direct feedback on several items in the form of a rating, the second is to disable the recommendation system until the user provides a minimal amount of feedback.

[0006] However, the number of requests per user is often less than or below the minimum amount of feedback, which can result in the system always being stuck in a cold start state, especially when the system is deployed on an aircraft or other vehicle as shown in FIG.

[0007] All publications identified in this specification are incorporated by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference. In the event that a definition or use of a term in an incorporated reference contradicts or is contrary to a definition of that term provided herein, the definition of that term provided herein applies and the definition of that term in the reference does not apply. Summary of the Invention [Problem to be solved by the invention]

[0008] Therefore, there remains a need for a system and method for a recommendation system that overcomes cold start situations in in-vehicle systems. [Means for solving the problem]

[0009] The present subject matter provides an apparatus, system, and method for providing a recommendation system for a vehicle content delivery network. A server having a processor and memory can be provided, communicatively coupled to an in-vehicle network for delivering content to multiple users. The server can be connected to multiple in-flight entertainment devices, typically located on the backs of the vehicle seats, but can also be connected to one or more user devices. Such user devices can include, for example, smartphones, tablet PCs, laptops, and other portable computing devices.

[0010] The static recommendation list can be generated based on trip characteristics stored in memory. The efficiency threshold can be calculated or provided to establish a point at which static recommendations are no longer used for recommendation information and instead a recommendation system is used. The efficiency threshold can be based at least in part on flight characteristics stored in memory, which can include, for example, (i) the length of the flight, (ii) the amount of content available on the flight, (iii) the type and / or variety of content available on the flight, and (iv) feedback from passengers on previous flights.

[0011] Once data regarding each user's preferences is collected based on implicit feedback, the data can be analyzed to calculate an efficiency level of the recommendation system operating for a user, and when the efficiency level is equal to or exceeds an efficiency threshold, the system can automatically switch from using a static recommendation list to using a recommendation system for that user.

[0012] Possible implicit feedback may include a user's interaction with the in-flight entertainment system, such as (i) a user selecting a piece of content, (ii) a user reviewing details of a piece of content, (iii) a user liking a piece of content, (iv) a user adding a piece of content to a playlist, and (v) a user skipping a piece of content.

[0013] Various objects, features, aspects and advantages of the present subject matter will become more apparent from the following detailed description of the preferred embodiments and the accompanying drawings, in which like numerals represent like elements. [Brief description of the drawings]

[0014] [Figure 1] 1 is an exemplary chart showing the number of movies watched per passenger during a flight. [Diagram 2] 1 is a flow chart of one embodiment of a method for providing a recommendation system for a vehicular content delivery network. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] Throughout the following description, numerous references are made to a server, service, interface, portal, platform, or other system formed from multiple computing devices. It should be understood that use of such terms is deemed to refer to one or more computing devices having at least one processor configured to execute software instructions stored on a computer-readable tangible, non-transitory medium. For example, a server can include one or more computers operating as a web server, database server, or other type of computer server performing the roles, responsibilities, or functions described.

[0016] The following discussion provides a number of exemplary embodiments of the subject matter of the present invention. Although each embodiment represents a single combination of the elements of the present invention, the subject matter of the present invention is considered to include all possible combinations of the elements disclosed. Thus, if one embodiment includes elements A, B, and C, and a second embodiment includes elements B and D, the subject matter of the present invention is considered to include other remaining combinations of A, B, C, or D, even if they are not explicitly disclosed.

[0017] Although the description herein focuses on aircraft, it is contemplated that the systems and methods described herein may be used on ships, trains, buses, and other vehicles as well.

[0018] The present subject matter describes systems and methods for providing a recommendation system that overcomes cold start situations, especially in situations where the number of user requests may be persistently low. Such systems and methods can be used to provide recommendation systems for in-vehicle networks where users have limited time to use the system, often on the order of a few hours.

[0019] The possible system and method consists of a series of steps. Figure 2 shows an embodiment of a method 200 for providing a recommendation system for a vehicle content delivery network. In step 205, a static recommendation list may be generated using a processor of the system or may be imported from an external system. The recommendation list is preferably based on a set of trip characteristics, which may include, for example, demographic information of the user, if possible. This recommendation list can be used in place of the recommendation system in a cold start phase. This may include, for example, the initial part of the flight, when the user has not yet interacted with the system.

[0020] In step 210, the system receives or calculates, for example, efficiency thresholds at which the recommendation system is expected to outperform a static recommendation list on a particular flight. Such thresholds may be computed dynamically during the flight or may be generated offline. It is contemplated that such thresholds may be specific to each flight and may vary between flights depending on various characteristics of the flight, such as the length of the flight, the amount of content available on the flight, the type and / or variety of content available on the flight, and feedback from passengers on previous flights.

[0021] In step 215, the system may generate implicit feedback using a processor each time a user makes a request to the system. If the generated implicit feedback is equal to or greater than the efficiency threshold for the flight, then in step 220, the system automatically enables the recommendation system for the user. Thus, for a particular flight, passengers or users may be grouped into two groups: (i) a group that uses the static recommendation list; and (ii) a group that uses the recommendation system. During the flight, some or all of the passengers or users may transition from the static recommendation list to the recommendation system when the implicit feedback to the users is equal to or greater than the efficiency threshold for the flight.

[0022] The system can also be configured to suggest various ways to solicit implicit feedback from the user, including asking the user to state whether they like or dislike a piece of content or a genre / category of content. Implicit feedback can also be collected in step 217 as the user interacts with the system. Such feedback can include, for example, which parts of the content were accessed or viewed by the user, and whether the user requested additional information about a piece of content (e.g., the user may read a more detailed description of a movie or watch a trailer, but end up not watching the movie). However, in this case, feedback can be provided in that regard, indicating an interest in a genre, actor, or a particular other type of movie, as compared to movies that have not been reviewed in detail.

[0023] Other possible feedback may include, for example, content that was skipped or not selected by the user, especially after details about the content have been reviewed, as well as playlists of content created by the user.

[0024] It is also contemplated that such a system could import a user's existing ratings, user profile, etc., if permitted by the user. This might include movie, television, etc. rankings from other services, which may be provided via applications stored on the smartphone, tablet PC, or other portable computing device. In such situations, it is contemplated that the existing information could be synchronized with the system.

[0025] As used herein, unless the context dictates otherwise, the term "coupled" is intended to include both direct coupling (where the two elements that are coupled to each other are in contact with each other) and indirect coupling (where at least one additional element is disposed between the two elements). Thus, the terms "coupled to" and "coupled with" are used synonymously.

[0026] In some embodiments, the numbers expressing properties such as amounts of ingredients, concentrations, reaction conditions, etc., used to describe and claim certain embodiments of the present invention should be understood to be modified in some cases by the term "about". Thus, in some embodiments, the numerical parameters set forth in the written description and accompanying claims are approximations that may vary depending on the desired properties sought to be obtained by a particular embodiment. In some embodiments, the numerical parameters should be construed in light of the number of reported significant digits and by applying ordinary rounding techniques. Notwithstanding that the numerical ranges and parameters setting forth the broad scope of some embodiments of the present invention are approximations, the numerical values ​​set forth in the specific examples are reported as precisely as practicable. The numerical values ​​presented in some embodiments of the present invention may contain certain errors necessarily resulting from the standard deviation found in their respective testing measurements.

[0027] Unless the context indicates a contrary intention, all ranges set forth herein should be construed as inclusive of their boundaries, and ranges without boundaries should be construed as including only those values ​​that are commercially practical. Similarly, unless the context indicates a contrary intention, all lists of values ​​should be considered to include intermediate values.

[0028] As used throughout this description and the claims that follow, the singular includes the plural unless the context clearly dictates otherwise. Also, as used in this description, the meaning of "in" includes the meanings of "in" and "on," unless the context clearly dictates otherwise.

[0029] The recitation of ranges of values ​​herein is intended to serve as a shorthand method of individually referencing each value within the range. Unless otherwise stated herein, each value with a range is incorporated herein as if it were individually stated herein. All methods described herein can be performed in any suitable order unless otherwise indicated herein or clearly contradicted by context. The use of any and all examples or exemplary language (e.g., "etc.") provided with respect to specific embodiments herein is intended merely to make the invention easier to understand and does not impose limitations on the scope of the invention as otherwise claimed. No language in this specification should be construed as indicating any non-claimed element essential to the practice of the invention.

[0030] Groupings of alternative elements or embodiments of the invention disclosed herein should not be construed as limitations. The elements of each group may be referenced and claimed individually or in any combination with other elements of the group or other elements found herein. One or more elements of a group may be included in or deleted from a group for reasons of convenience and / or patentability. When such inclusion or deletion occurs, the specification is deemed to include the modified group and thus fulfills all Markush-style group descriptions used in the appended claims.

[0031] It should be apparent to one skilled in the art that many more modifications beyond those already described are possible without departing from the inventive concept herein. Accordingly, the subject matter of the present invention should not be limited except in the spirit of the appended claims. Moreover, in interpreting both the specification and the claims, all terms should be interpreted in the broadest possible manner consistent with the context. In particular, the terms "comprise" and "include" should be interpreted as referring to elements, configurations, or steps in a non-exclusive manner, indicating that the referenced element, configuration, or step may be present, utilized, or combined with other elements, configurations, or steps not expressly referenced. If a claim of the specification refers to at least one selected from the group consisting of A, B, C···N, the sentence should be interpreted as requiring only one element from the group, not A and N, or B and N, etc.

Claims

1. 1. A method for providing a recommendation system for a vehicular content delivery network, comprising: providing a server for an entertainment system, the server having a processor and memory and communicatively coupled to an in-vehicle network of a plurality of entertainment devices for delivering content to a plurality of users; generating, using the processor, a static recommendation list of the content based on the travel characteristics stored in the memory; receiving or calculating an efficiency threshold based at least in part on the trip characteristics stored in the memory; displaying, for each of the plurality of users, content suggested to the user based on the static recommendation list on a display specific to the user, the display being a display of one of the plurality of entertainment devices; collecting implicit feedback data regarding each user's preferences, the implicit feedback data including the user's interactions with one of the plurality of entertainment devices via the display, the user's interactions including one or more of: (i) the user selecting a piece of content; (ii) the user reviewing details of the piece of content; (iii) the user liking a piece of content; (iv) the user adding a piece of content to a playlist; and (v) the user skipping a piece of content; analyzing the collected implicit feedback data using the processor for each of the plurality of users to calculate an efficiency level of the recommendation system based on the collected implicit feedback of that user; comparing the efficiency level to an efficiency threshold to determine whether the efficiency level is greater than or equal to the efficiency threshold, and when the efficiency level is greater than or equal to the efficiency threshold, automatically switching from the static recommendation list to use of a recommendation system for the user; upon switching to a recommendation system for the user, updating the suggested content on the display for that user based on the recommendation system for the user; A method comprising:

2. The recommendation system is deployed on an aircraft, the efficiency threshold is based at least in part on flight characteristics of a flight of the aircraft; the flight characteristics include at least one of: (i) the length of the flight; (ii) the amount of content available on the flight; (iii) the type and / or variety of content available on the flight; and (iv) feedback from passengers on previous flights; The method of claim 1.

3. The recommendation system is deployed on an aircraft, the efficiency thresholds vary for different flights of the aircraft; The method of claim 1.

4. implementing the static recommendation list in a cold start phase, prior to collecting data regarding each user's preferences. The method of claim 1.

Images