WO2016049434A1

WO2016049434A1 - Systems and methods for state connections

Info

Publication number: WO2016049434A1
Application number: PCT/US2015/052185
Authority: WO
Inventors: Alexander Asseily; Mark L. ASSEILY; Nicholas W. R. JAKOBI; Benjamin MILES; Jiten BHAGAT; Daniel Harvey; Alexander NAKONECHNYJ
Original assignee: EQUAL MEDIA Ltd
Current assignee: EQUAL MEDIA Ltd
Priority date: 2014-09-26
Filing date: 2015-09-25
Publication date: 2016-03-31
Anticipated expiration: 2017-03-26
Also published as: GB201417092D0

Abstract

A recommendation system and method for connecting unique users of an opinion network with interactive opinion flows by structuring and aggregating electronic opinion data using a sentiment-based or other collaborative filter. The recommendation system permits the distribution of user published content (i.e., opinions) to a receptive and equitable audience, who can in turn respond to and engage with content to which they may not otherwise have access due to the asymmetric distribution of online and offline media.

Description

S P E C I F I C A T I O N

SYSTEMS AND METHODS FOR STATE CONNECTIONS

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to UK Patent Application Number 1417092.2, which was filed September 26, 2014. The disclosure of the UK application is herein incorporated by reference in its entirety and for all purposes.

FIELD

[1] The present disclosure relates generally to systems and methods for managing opinion networks with interactive opinion flows and more particularly, but not exclusively, to systems and methods for electronic recommendation systems in an opinion network.

BACKGROUND

[2] Social networks connect users to other users that they have come into contact with in other online and offline environments. A user manually 'follows' or adds other users to their network, limiting the scope and size of a user's audience to people in their existing networks. Audience reach is, therefore, static or inherently tied to a user's popularity within their existing online and offline networks. Whilst the act of 'adding,' 'liking,' or 'following' users on social networks makes published content accessible to, and distributable by, the user, this content is typically disseminated across asymmetric, one-to-one, and one-to-many channels.

[3] In view of the foregoing, a need exists for a recommendation system to automatically connect users of an opinion network that do not previously know each another based on their opinions (and by extension, interests), to create unique audiences for users of social networks that are not limited to the user's existing social connections. A further need exists to permit the distribution of user published content (i.e., opinions) to a receptive and equitable audience, who can in turn respond to and engage with content to which they may not otherwise have access due to the asymmetric distribution of online and offline media. The present disclosure relates to such a recommendation system that fulfils the value proposition, "connect to people based on what you think, not who you know. " SUMMARY

[4] The present disclosure relates generally to a recommendation system, method and algorithm for connecting unique users of an opinion network with interactive opinion flows by structuring and aggregating electronic opinion data using a sentiment-based or other collaborative filter. Example opinion networks are further disclosed in United States Patent Application Serial No. 14/238,193, filed on August 14, 2012, entitled "System and Method for Managing Opinion Networks with Interactive Opinion Flows." The disclosure of United States Patent Application Serial No. 14/238,193 is hereby incorporated by reference in its entireties and for all purposes.

[5] Each time a user publishes an opinion on the system, the user is automatically connected to one or more users. Connections are based in the first order on agreement, or any other given variable (see, e.g., Fig. 6). The act of stating in and of itself connects a user to other members of an implicit group that are also discussing the same topic, or entity (a "conversation"). The more frequently a user states an opinion, the richer the dimensions of connectivity with other users. The recommendation system produces unique lists of connections that are global and/or distinguished by topic category, and which can be located in the user's profile page ("Connections") or in an "Explore" page on a mobile application.

[6] More particularly, but not exclusively, this disclosure relates to the end-to-end user experience of publishing x opinions, where x > 1, to generate n connections, where 0 < n < number of users on opinion network with x opinions, to whom to distribute a user's statement results. A user's opinions are structured and aggregated in categories, subcategories and subregions in the user's profile page (collectively, their 'networks'), and distributed in the connected user's "For You" stream, thereby enabling a user to "Z?e heard . In another embodiment, connections may directly or indirectly inform and/or engender a discovery process between two connected users, permitting them to be in direct contact.

[7] In one embodiment, the presently disclosed systems and methods are applicable to all operating systems (e.g., iOS®, Android®, Windows® phone, Firefox® OS and others), Web platforms (e.g., website, application programming interfaces (APIs), widgets, embedded applications, and so on), and desktop applications (e.g., Mac® OS, Linux®, and Windows®).

[8] This summary is provided to introduce the subject matter of the disclosure and not intended to identify essential features of the claimed subject matter, nor is it intended for use in determining the scope of the claimed subject matter. Other systems, methods, features and advantages of the disclosure will be or will become apparent to one with skill in the art upon examination of the following figures and detailed description. It is intended that all such additional systems, methods, features, and advantages be included within this description, be within the scope of the disclosure, and be protected by the accompanying claims.

BRIEF DESCRIPTION OF THE DIAGRAMS

[9] In order to better appreciate how the above-recited and other advantages and objects of the disclosure are obtained, a more particular description of the embodiments briefly described above will be rendered by reference to specific embodiments thereof, which are illustrated in the accompanying diagrams. It should be noted that the components in the figures are not necessarily to scale, emphasis instead being placed upon illustrating principles of the disclosure. Moreover, in the figures, like reference numerals designate corresponding parts throughout the different views. However, like parts do not always have like reference numerals. Moreover, all illustrations are intended to convey concepts, where relative sizes, shapes, and other detailed attributes may be illustrated schematically rather than literally or precisely.

[10] Figs. 1 - 5 are schematic diagrams depicting aspects of an example graphical mobile user interface for building connections in an interactive opinion network flow, in accordance with at least one embodiment of the disclosure.

[11] Figs. 6 - 8 are schematic diagrams depicting aspects of an example graphical mobile user interface for profiling a user's statement results and connections in aggregate domains, in accordance with at least one embodiment of the disclosure.

[12] Figs 9 - 10 are schematic diagrams depicting aspects of an example graphical mobile user interface for profiling domain and statement overlaps between two connected users.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[13] Fig. 1 Landing page after sign-up. Prompts user to state opinion (must state > 1 opinions to build connections) by clicking an 'Explore' button.

[14] Fig. 2 Opinion statement. User selects up to a predetermined number (e.g., three) opinion words from a word map. Electronic input data includes words of a natural language, sentence fragments of natural language, sentences, or graphics/video/audio corresponding to words of a natural language.

[15] Fig. 3 Statement results. Opinion is published and 'connected' to other opinions in domain-based networks (e.g. , categories) and to other users in implicit agreement on the respective entity (i.e., connections). In one embodiment, agreement is captured in semantic meaning of opinion word and related words (e.g., synonyms, antonyms, hypernyms) or derived from links to a lexical database, which semantically clusters concepts and relates them to a hypernym taxonomy. This may reference one or more synsets - a group of opinion words that are synonymous or have sufficiently similar meaning. 'Agreement' is also captured in sentiment polarity of opinion itself - as compared to the individual opinion word - taking into account negation and modifiers, prior-polarity tags, and the entity it relates to. Statement results syntax, "81 people agree with you, " refers to the community of users in implicit agreement on a given entity. The diagram of a world map in Fig. 3 shows a geographical location of all users in agreement towards a given entity. In one embodiment, geographical locations for each user can be determined using an Internet Protocol (IP) address of the particular user. As each opinion is aggregated into domain networks, user sees graphical notification in the user interface notifying them that their opinions have been counted and published. This notification includes a link to the location of the published aggregate opinion and map showing the unique geographical locations of a user's new connections.

[16] Fig. 4 Statement results. All particular reasons that have been expressed for a given opinion, including statements that structurally link unrelated opinions/entities and long form statements. This is shown as a stream of statement results whose sentiment polarity correlates with the user's original opinion. All statements in this stream belong to a user's new connections. Provides links to the location of published aggregate opinion (entity page) and to connected users' profiles. The user can respond to, comment on, or 'Well Said' individual statement results from connections.

[17] Fig. 5 "For You" stream is automatically populated with opinion data from newly formed connections and/ or by domain overlap. Notification syntax: "You are now connected. " Statements are distributed in the "For You" stream by a relevance ranking: i) most recent statements made by new connections with domain overlap, ii) most recent statements on trending entities based on domain recognition of category or subcategory in which the user has opined (including statements made by other users that are not yet 'connections') iii) long form statements or statements with a high 'Well Said' count. As a user's number of connections grows, the statement results distributed in the user's For You stream will be biased towards the user's closest connections, i.e., connections with the strongest opinion sentiment overlaps. Overlaps will count less if the overlap is with an old opinion or with an opinion made many opinions ago by another user. This is to prevent old users, users that are no longer active on the network, and/or users that have published many opinions from dominating a user's top connections list and "For You" stream.

[18] In another embodiment, there are variable ways of ordering the relevance ranking for the connections recommendation system. For example, the opinions distributed in a user's stream may originate from trusted sources, i.e., their connections, but are not filtered by a bias towards implicit agreement where a domain, category or subcategory overlap already exists. This permits users' opinions to be 'challenged' by their closest connections or by those most receptive to a user's opinions. In another embodiment, relevance ranking can be applied to geography or any other applicable trait.

[19] Fig. 6 Profile page. User's opinions are aggregated by domain. World map shows unique geographical locations of a user's connections.

[20] Fig. 7 User's opinions aggregated by domain. Structuring opinion data by category or subcategory will provide the user with an indication of their relative audience in the respective domain. User can infer that their number of connections will grow as their opinion count grows by quantitative summary of statement results overall and by category. Users can dive into any category to retrieve their statement results in the respective category.

[21] Fig. 8 Connections list by category. A user sees a list of connections within a category in which both the user and the user's connection/s has/have published one or more opinions, ranked in order of 'agreement' or sentiment polarity. In one embodiment, a connection score between two users is calculated based on the number of topics that overlap and "agree" upon, and the activity of each user. This connection score can be used to rank connections. This is illustrated in the user interface by the placement of two congruous diamonds figurative of a user's margin of agreement with each of their unique connections. The greater the sentiment overlap, the greater the overlap of the diamonds. The plain text underneath the connections' avatars indicates the statement within the respective domain on which there is strongest agreement between two connected users. [22] Fig. 9 Connections profile page. Shows connections overlap by placement of diamond avatars, which allows two connected users to see how connected they are at a glance. Overlap can be incorporated as a feature of the avatar itself or as a separate embodiment. A user can manually "subscribe" to another user, which overrides the recommendation system filter. If a user subscribes to another user in their existing connections list, the override will distribute all statement results across all domains, categories, subcategories, etc. from the subscribed user into a user's For You stream, regardless of domain overlap or lack thereof. Conversely, if a user "unsubscribes" from another user in their existing connections list, all statement results from the unsubscribed user will be removed from the user's For You stream, regardless of their prior 'connectivity'. This permits users to exercise some level of manual control over the statement results published in their stream, and to further the idea of building unique audiences that are receptive around a user's opinions.

[23] Non-illustrated embodiment. Distribution of a user's statement results in the streams of other users, using the same or some variant of the relevance ranking described in Paras. [0018] and [0019] above. Non-illustrated embodiment. Aggregated opinions from a user's connections may be used as an implied recommendation system for products and service, and such recommendations can be monetised as a result.

Claims

CLAIMS What is claimed is:

1. A method of connecting users of an opinion network, comprising: receiving a selection of topics for stating an opinion;

receiving a selection of predetermined opinion words from an opinion map that represents the opinion regarding said received selection of topics; publishing the received selection of predetermined opinion words; and connecting one or more users based on the published selection of predetermined opinion words,

wherein each of said predetermined opinion words is defined by a sentiment polarity and said connecting the one or more users is based on an agreement in sentiment polarity of said published selection.

2. The method of claim I, wherein said connecting is further based on a geographical location of said one or more users.

3. The method of claim 1, further comprising providing links to profiles of said connected users.

4. The method of claim 1, further comprising providing opinions of said connected users.

5. The method of claim 4, wherein said providing opinions of said connected users includes ranking said opinions based on a connection score between said connected users.