CN118103831A - Automatic identification of additional content of a web page - Google Patents
Automatic identification of additional content of a web page Download PDFInfo
- Publication number
- CN118103831A CN118103831A CN202280068964.2A CN202280068964A CN118103831A CN 118103831 A CN118103831 A CN 118103831A CN 202280068964 A CN202280068964 A CN 202280068964A CN 118103831 A CN118103831 A CN 118103831A
- Authority
- CN
- China
- Prior art keywords
- web page
- additional content
- side pane
- content
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 claims abstract description 126
- 238000010801 machine learning Methods 0.000 claims description 36
- 238000010411 cooking Methods 0.000 claims description 25
- 230000003993 interaction Effects 0.000 claims description 22
- 230000004044 response Effects 0.000 claims description 21
- 238000005457 optimization Methods 0.000 claims description 18
- 238000009877 rendering Methods 0.000 claims description 15
- 230000004931 aggregating effect Effects 0.000 claims description 11
- 230000001960 triggered effect Effects 0.000 claims description 7
- 238000013145 classification model Methods 0.000 claims description 5
- 239000013589 supplement Substances 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 10
- 238000003860 storage Methods 0.000 description 9
- 230000009471 action Effects 0.000 description 8
- 239000000047 product Substances 0.000 description 8
- 230000008901 benefit Effects 0.000 description 6
- 230000002452 interceptive effect Effects 0.000 description 4
- 239000002994 raw material Substances 0.000 description 4
- 241000700605 Viruses Species 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 230000003247 decreasing effect Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000010006 flight Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000003306 harvesting Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 229920001690 polydopamine Polymers 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000008685 targeting Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Landscapes
- Information Transfer Between Computers (AREA)
Abstract
The present disclosure relates to systems and methods for automatically identifying additional content of a web page. The system and method may create a side pane that is placed next to the web page to supplement the web page in the browser. The side pane may provide additional information or additional content to assist the user in consuming the web page. The additional content may provide web page insight and help the user further explore the web page.
Description
Background
The user interacts with the browser for content discovery, for example by viewing different web pages or viewing multimedia content (photos, videos, images). The user also interacts with the browser to perform end-to-end task completion (e.g., executing a search query for a topic and looking up related content or answers for the topic, or searching for a product and purchasing a product). Currently, when a user is viewing a web page on a browser, the browser does not provide a solution to help the user understand the current web page content.
Disclosure of Invention
The present disclosure describes systems and methods for identifying additional content of a web page. For example, in response to a user accessing a web page having an article about a virus, the system may identify additional content to present in a side pane in the vicinity of the web page. The additional content may include a summary of the article and a gist of the article such that a user may read the gist of the article to obtain an understanding of the web page content. The side pane may remain in view of a web page with an article about the virus. As described in more detail below, additional content may be identified based on the content or context of the web page that the user is accessing. The additional content may also include content derived from the web page (e.g., information from the article). In this way, the system presents additional content in a side pane that helps the user consume the web page or supplement the web page.
This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.
One example implementation relates to a method for automatically identifying additional content of a web page. The method may include receiving a request for additional content of a web page, wherein the request includes a Uniform Resource Locator (URL) of the web page. The method may include categorizing the URL of the web page. The method may include triggering a side pane experience for additional content based on the classification of the web page. The method may include aggregating additional content obtained from one or more data sources for a side pane experience, wherein the additional content is obtained based on a context of the web page. The context of a web page includes the content of the web page. The method may include sending the additional content to a browser to be presented in a side pane near the web page. The method improves user efficiency by organizing and arranging additional content into modules. In addition, the method reduces the need for the user to navigate to multiple websites and thus reduces user input and/or clicks.
Another example implementation relates to a method for presenting additional content of a web page. The method may include identifying a Uniform Resource Locator (URL) of a web page accessed by a browser. The method may include sending a request for additional content, wherein the request includes a URL of a web page. The method may include receiving additional content, wherein the additional content is based on a context of the web page. The method may include presenting the additional content in a side pane beside the web page.
Additional features and advantages will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the teachings herein. The features and advantages of the disclosure may be realized and obtained by means of the instruments and combinations particularly pointed out in the appended claims. The features of the present disclosure will become more fully apparent from the following description and appended claims, or may be learned by the practice of the disclosure as set forth hereinafter.
Drawings
In order to describe the manner in which the above-recited and other features of the disclosure can be obtained, a more particular description will be rendered by reference to specific implementations thereof which are illustrated in the appended drawings. For better understanding, like elements are designated by like reference numerals throughout the various figures. Although some of the drawings may be shown in schematic or exaggerated representations of concepts, at least some of the drawings may be drawn to scale. It is to be understood that the drawings depict some example implementations that will be described and explained with additional specificity and detail through the use of the accompanying drawings in which:
FIG. 1 illustrates an example environment for identifying and rendering additional content of a web page in accordance with implementations of the present disclosure.
FIG. 2 illustrates an example method for identifying additional content of a web page according to an implementation of the disclosure.
FIG. 3 illustrates an example graphical user interface of a browser displaying a web page and a side pane including a summary module according to an implementation of the disclosure.
FIG. 4 illustrates an example graphical user interface of a browser displaying a web page and a side pane including a page overview module and a theme module, according to an implementation of the present disclosure.
FIG. 5 illustrates an example graphical user interface of a browser displaying a web page and a side pane including a question and answer module according to an implementation of the present disclosure.
FIG. 6 illustrates an example graphical user interface of a browser displaying a web page and a side pane including a knowledge card module, in accordance with an implementation of the present disclosure.
FIG. 7 illustrates an example graphical user interface of a browser displaying a web page and including a side pane with respect to a source module according to an implementation of the present disclosure.
FIG. 8 illustrates an example graphical user interface of a browser displaying a web page and a side pane including a recommended content module according to an implementation of the disclosure.
FIG. 9 illustrates an example graphical user interface of a browser displaying a web page and a side pane including a book module according to an implementation of the disclosure.
FIG. 10 illustrates an example graphical user interface including a side pane of a shopping module according to an implementation of the present disclosure.
FIG. 11 illustrates an example graphical user interface including a side pane of a travel module according to an implementation of the disclosure.
Fig. 12 illustrates an example graphical user interface including a side pane of a food module according to an implementation of the present disclosure.
Fig. 13 illustrates an example graphical user interface of a browser displaying a web page and a side pane including a cooking module according to an implementation of the present disclosure.
FIG. 14 illustrates an example graphical user interface of a browser displaying a web page and a side pane including related videos and images according to an implementation of the present disclosure.
FIG. 15 illustrates an example method for automatically identifying additional content of a web page according to an implementation of the present disclosure.
FIG. 16 illustrates an example method for presenting additional content of a web page according to an implementation of the disclosure.
Detailed description of the invention
The present disclosure relates generally to identifying additional content for a web page that a user is currently viewing or interacting with. The user interacts with the browser for content discovery, for example by viewing different web pages or viewing multimedia content (photos, videos, images). The user also interacts with the browser to perform end-to-end task completion (e.g., executing a search query for a topic and looking up related content or answers for the topic, or searching for a product and purchasing a product). Currently, when a user is viewing a web page on a browser, the browser does not provide a solution that helps the user understand the current web page content.
The system and method of the present disclosure helps a user complete end-to-end tasks while browsing a wide variety of web pages without disrupting the user's current workflow. The present invention provides web page insights, such as web page summaries, web page topics, identifying or providing key entities that assist a user in consuming current web page content, and/or recommending context or personalized content (e.g., related articles for further exploration by the user).
The systems and methods of the present disclosure provide a user with a contextual and personalized feed having additional content related to the content the user is currently consuming. The present disclosure helps users better understand the currently viewed content and helps users explore related content. In addition, the present disclosure enables users to interact with content being viewed or with users having similar interests.
The systems and methods of the present disclosure create a side pane that may be placed on the right side of the browser to supplement the web page that the user is viewing. The side pane provides additional information and/or additional content to the user to help the user consume the current web page that the user is browsing using the browser. The side panes remain in the front view of the web page.
The side pane may be positioned in the right column of the browser with a contextualized and/or personalized feed for additional content. As an companion canvas to the web page that the user is currently browsing, the side panes are opened either reactively by the user or actively by the algorithm. Contextualized and/or personalized summaries may be organized into different tabs or modules.
One example use case is for a user to view a web page for a book. Example tabs or modules shown in the side pane are included on this page module, which shows the shorthand as a web page summary, followed by a list of salient topics, entities and key-phrases extracted from the web page as its overview, common questions (FAQ) about the web page content, and the number of web page views and social reactions (number of likes, superloves, surprise, etc.) to show the popularity of the web page. Since the web pages are about the book, the present disclosure may also present rich knowledge cards of the book in a side pane so that interested users may learn more about the book, read their comments, and/or purchase the book. Another example tab or module shown in the side pane includes a module that recommends you more of related articles, a summary of questions and answers (e.g., about books in web pages, authors, her census, climate change, etc.), related videos/images (such as book previews, cover images, etc.), related entities (other books of authors, about climate change, summer reading list, etc.), and link(s) or suggestion query(s) for related searches to explore more for users. In some implementations, selecting a link or suggesting a query results in the corresponding search results being displayed in a side window. Another example table or module shown in the side pane includes a "about origin" module that provides a listing of a website description, website quality (e.g., quality in terms of spam, unwanted mail, adult mail, or malware detection) and similar websites for magazines of web pages published on a book. The side pane may also show topic authority information for the website, user traffic trends for the web page, and their incoming links to help the user assess the trustworthiness of the source. Another example table or module shown in the side pane includes a user-generated content module that allows a user to compose comments on a book, as well as ask and answer questions about the book and author.
As such, the systems and methods of the present disclosure provide web page insights, such as page summaries, page topics, and/or key entities, that help users consume the content of the current web page and recommend context or personalized content (related articles) for further exploration by the users. The personalized content is based on the context in which the user is browsing to help maintain the user interactive browsing session. Personalized content may also include user-generated content (e.g., comments on a web page).
The systems and methods of the present disclosure may reactively open a side pane based on user input. The user may click on a toolbar icon on the browser and, in response to the user clicking on the toolbar icon, a side window with additional content (e.g., recommended context or personalized content) may be opened. The toolbar icon may be enabled at all times. As such, the user may be able to click on the toolbar icon whenever the user is browsing a web page using the browser.
The additional content presented in the side pane may be based on the context of the web page that the user is currently browsing. The context of a web page includes the content of the web page. The system and method searches for relevant material and optionally filters out results that are identical to those already present in the web page content. In addition, the additional content may include content derived from a web page. For example, one or more machine learning models may identify content on a web page and present the content of the web page as additional content in a side pane.
The side pane may have one or more modules highlighting the additional content. The side pane's modules may help users consume content, help users explore additional content, and/or help maintain a user interactive browsing experience. Additional content presented in the side pane may come from multiple sources. For example, if a web page has a primary entity (e.g., seattle hawk team), entity cards for the primary entity may be presented in a side pane with content syndication from multiple sources.
The modules on the side pane may include default modules (e.g., overview module, source module, explorer module). The overview module may include a number of modules that assist the user in consuming the web page (e.g., points in the web page, web page topics, questions, and answers). The overview module may also have a view more button in which the user may view additional content related to the web page.
The gist module of the side pane may assist the user in reviewing the gist harvest from articles on the web page. In addition, if the user compares the time of the approach, the user can skip reading the entire content of the web page and go directly to the gist to understand the content of the web page.
The web page topic module of the side pane may be generated by a machine learning model reviewing content of the web page and associating the content of the web page with or relating to topics referred to in the web page. The user may click on the topic and perform a search for the selected topic in the side pane (e.g., phrases for the selected topic are automatically entered as search terms and perform a search for the selected topic). The search results are provided in a side pane, and the user can browse the search results for the selected topic in the side pane while maintaining the web page that the user is currently browsing in the left pane.
The side pane question and answer module may include a plurality of questions generated by a machine learning model having answers provided by content provided by a web page.
The side pane may also have a exploration module that helps the user continue exploration and inspiring based on the content of the web page. The exploration module may provide content recommendations related to a current web page based on the context of the current web page. The personalized content module may personalize the provided content based on the user's behavior and/or the user's past interactions with the browser.
If the web page is private (e.g., a bank login page) and the machine learning model cannot identify the content of the web page, the exploration module may present personalized content based on the behavior of the user in the past interactions with the browser. User interaction with the browser may include, for example, selecting a banner on a web page, a web page viewed, and/or a previous search topic. The presented content may be personalized based on previous search interests (sports, health, do by oneself (DIY)).
The side pane may also have a "about origin" module that provides information about the origin of the web page (disclaimer of origin, description of origin, quick links that the user can continue browsing).
The modules included in the side panes may vary based on the domain of the web page. For example, news, finance, shopping, weather, maps, sports, shopping, cooking, and travel domains may include different modules in the side panes. One example module for a sports web page or news web page includes an entity module. If the web page has a primary entity (e.g., seattle hawk team), the entity card for the primary entity may be presented in a side pane with content syndication from multiple sources.
Another example module for a travel web page may include a exploration module that allows a user to explore contextual recommendations for a place (e.g., hotels for a place, searching for flights for places, car rental recommendations, recommendations for attractions or restaurants, and/or having a map of attractions or restaurants highlighting the place) based on the content of the web page the user is viewing or a search performed by the user.
Another example includes, for a cooking web page, modules including information about recipes (e.g., raw materials, cooking times, serving portions) mentioned in the cooking web page and/or modules having recipes related to recipes mentioned in the cooking web page.
The user can open any module of the side pane in the new tab. In addition, the user may adjust the size of the side pane by increasing and/or decreasing the size of the side pane.
In one implementation, the side panes are automatically presented. The side pane may be automatically presented based on the domain or category of the web page accessed by the user. For example, a subset of web pages of a Uniform Resource Locator (URL) (e.g., cooking, travel, sports) may automatically trigger a side pane to display in a browser near or next to a web page selected by a user.
The side pane may also include search input, where a user may enter search terms for a query and search results for the query may appear within the side pane. In this manner, a user may run multiple searches in the side pane while viewing a web page.
The system and method of the present disclosure may be used as a reading assistant and content discovery engine. The present disclosure utilizes natural language understanding, machine understanding, and personalized recommendation techniques to correlate content with a user's current tasks and long-term interests. In some implementations, the present disclosure is a companion browser with a mobile screen size that runs alongside the main browser to proactively provide relevant information and enable further exploration without requiring the user to be taken off of the current task workflow.
The system and method of the present disclosure can be easily accessed and provide content aggregated across the entire world wide web, no matter which website the user is browsing, which is an advantage over other content services that create data islands.
One technical advantage of some implementations of the systems and methods of the present disclosure is to provide additional content supported by a browser and, thus, may increase user interactivity with the browser. In addition, the system and method improves user efficiency by organizing and arranging additional content into modules. The system and method also reduces the need for a user to navigate to multiple websites and thus reduces user input and/or clicks. The system and method also improves the user's trust because several of the modules allow the user to understand the origin and reputation of the website. In addition, the system and method improves accessibility to websites that otherwise do not have good accessibility options.
As such, the systems and methods of the present disclosure provide companion canvas that gives the user insight into the currently viewed web documents, enabling deep exploration of current topics, suggestion of new topics, understanding of various perspectives, and/or providing other contextual experiences for the user.
Referring now to FIG. 1, an example environment 100 for identifying and presenting additional content of a web page is illustrated. The environment 100 may include one or more users 114 interacting with one or more browsers 102 on the device. Browser 102 allows user 114 to interact with information on the world wide web. When the user 114 requests the web page 30 from a web site (e.g., by performing a search using the browser 102 or entering a Uniform Resource Locator (URL) of the web site using the browser 102), the browser 102 retrieves the content of the web page 30 from the web server and displays the web page 30 on the display of the user device. The web page 30 may be any web page (a third party web page or a web page from the same party that provided the browser 102). In addition, browser 102 may be a browser application on the device of user 114. Examples of browser 102 include, but are not limited to, EDGE TM and INTERNET EXPLORER TM.
The browser 102 can have a user interface rendering component 10 that presents the requested web page 30 on a display. The user interface drawing component 10 may also present the side pane 26 with additional content 28 for the requested web page 30. Side pane 26 may be presented next to or near web page 30. For example, the side pane 26 is presented by the user interface rendering component 10 on the right side of the requested web page 30. In some implementations, the browser 102 provides a notification to the user 114 to inform the user 114 that additional content 28 for the requested web page 30 may be present that may be of interest to the user. The notification may identify an icon or button on the browser 102 that the user 114 may click to receive the side pane 26 with additional information.
The user 114 may select an icon or button on the browser 102 to trigger the presentation of the side pane 26 with additional information and/or additional content 28. The browser 102 can send a request 24 to the runtime server 104 for additional content 28 to be presented in the side pane 26, and the user interface drawing component 10 can present the side pane 26 with the received additional content 28. The browser 102 can send the request 24 for the additional content 28 and the user interface drawing component 10 can present the side pane 26 with the additional content 28 based on a reactive trigger (e.g., the user 114 clicking a button).
In addition, the browser 102 may automatically send the request 24 for additional content 28 to the runtime server 104, and the user interface rendering component 10 may automatically present the side pane 26 with the received additional content 28 on the browser 102 in response to the user opening the web page URL. Thus, the presentation of the side pane 26 with the additional content 28 on the browser 102 may be triggered reactively (e.g., by the user 114 selecting an icon or button) or proactively (e.g., automatically when the user opens a web page URL).
The runtime server 104 may be a search engine that aggregates and returns relevant and useful content about the web page 30 that the user 114 opened on the browser 102. The runtime server 104 provides a single endpoint of connection for the browser 102 by aggregating the different data sources 106 for the additional content 28 presented in the side pane 26. The runtime server 104 also organizes the entire page experience for the side pane 26 and handles interactions of the user 114 with the side pane 26.
The input to the runtime server 104 is different from a conventional search engine, where the input from the user 114 is a query. The runtime server 104 receives as input the URL of the web page 30 requested by the user 114 and may also receive as input user interactions (e.g., selecting a topic of interest to view additional content) on the side pane 26.
The runtime server 104 can include an event dispatcher component 12 that receives a request 24 from the browser 102 with a URL of a web page 30. Event dispatcher component 12 determines whether request 24 is a new request (e.g., a new web page opened by user 114) or whether request 24 is a user interaction of user 114 with side pane 26.
If the request 24 is a new request (e.g., for a newly accessed web page), the event dispatcher component 12 sends the request 24 with the URL of the web page 30 to the page classification component 14. The page classification component 14 runs URL classification of the URL of the web page 30 and determines whether to trigger the side pane 26 based on the URL classification. Whether the side pane 26 is triggered may be a rule-based trigger (e.g., a URL or domain-based list trigger) or a custom trigger (e.g., a partner service-based trigger). The URL classification may determine whether the URL of the web page 30 is a private domain or a public domain. In addition, the URL classifications determine the domain of the web page 30 (e.g., sports, travel, cooking, news, medical, etc.).
If the classification identifies the URL as a private webpage 30, the page classification component 14 may block presentation of the side pane 26 with additional content based on the private webpage classification. Additionally, if the classification identifies a URL that is a domain included on the domain list to block or prevent presentation of the side pane 26 (e.g., offensive domain, adult content domain, spam domain), the page classification component 14 may block presentation of the side pane 26 with the additional content 28 based on the domain of the web page 30. The page classification component 14 can have one or more machine learning models that perform URL classification and determine whether to trigger presentation of the side pane 26 with the additional content 28. As such, the page classification component 14 can determine whether to trigger the side pane 26 with the additional content 28 based on the classification of the URL of the web page 30.
The page classification component 14 can communicate with the experience triggering component 16 to determine side pane experiences for the additional content 28 based on the classification. Experience triggering component 16 may determine one or more modules and/or experience levels to provide additional content 28 in side pane 26. Different modules and/or experiences may be presented based on the classification of web pages 30. For example, experience trigger component 16 may select different modules for different web page domains.
The selection of the side pane experience may be a rule-based trigger (e.g., trigger selection of a different module based on a list of URLs or domains). The generic experience may be triggered for the web page 30 (e.g., a set of predefined modules may be presented in the side pane) or the domain-specific experience may be triggered based on the web page 30 (e.g., a particular set of modules is selected for presentation in the side pane). Example modules include, but are not limited to, a summary module, a question and answer module, a recommended content module, a personalized content module, a page overview module, an exploration module, a theme module, a point module, a knowledge card module, a source module, a cooking module, a food module, a travel module, a shopping module, and/or a book module.
For example, the knowledge card module trigger is determined by the dominant entity marking result for web page 30, and the page summary module trigger is based on whether the original web page is worth the summary and whether the summary will be helpful to end user 114. Another example for a cooking web page includes a module having information about recipes mentioned in the cooking web page (e.g., raw materials, cooking time, serving portion) and/or a module having recipes related to recipes mentioned in the cooking web page selected by experience triggering component 16. Another example of a private web page (e.g., a bank login page) includes an experience trigger component 16 that selects a module to present personalized content. Experience triggering component 16 may have one or more machine learning models that determine which modules or experience levels to use in side panes 26 of web page 30 and/or which additional content 28 to include in side panes 26. As such, experience triggering component 16 decides which side pane experience to trigger for input web page 30.
After the page classification component 14 and/or experience trigger component 16 make a trigger decision, the data generator component 18 issues a request for data collection to the corresponding data source 106. The data sources 106 may include different data sources from heterogeneous data providers of different content (e.g., article recommendations, questions, and answers). For example, the page summary may be from a document index. In addition, other additional content (e.g., knowledge cards, entity panes, and/or related answers) may be integrated by directly invoking different data sources 106. When a request arrives at the runtime server 104, the data source 106 may provide content at runtime.
The data aggregator component 20 aggregates content collected from the data sources 106 and aggregates content collected from the content data store 108. The content data store 108 may be a central document index store accessed by the data aggregator component 20 when a request 24 with a URL is received at the runtime server 104.
The content data store 108 can have one or more machine learning models that generate content offline. To support the capacity of web pages (e.g., important web pages or static web pages), the content data store 108 may receive content in an offline triggered batch mode over the web index data after the URL selection module. The content data store 108 may also receive content in a reactive streaming mode that is triggered when a user opens a web page that is not in the content store and/or web index. Content data store 108 may also receive content when a web index crawler crawls new web pages. Together, these three modes continue to update the content data store 108 to continue to refresh the stored content and provide the updated content to the data aggregator component 20 with side panes for use. The data aggregation component 20 provides greater flexibility to the runtime server 104 to handle different data sources 106.
The data aggregator component 20 stores the additional content 28 as a snapshot of the current user session for the web page 30 in the session data store 110. The snapshot provides an overview of the additional content 28 presented to the user in the side pane 26 in the user session for the web page 30. The snapshot may be used to update the additional content 28 in the side pane 26 in response to user interaction with the additional content 28 to prevent presentation of the same additional content 28 to the user 114. User data store 112 may store content related to user profiles of users 114.
Composer component 22 retrieves data (e.g., additional content) from session data store 110 and organizes additional content 28 for rendering. The orchestrator component 22 completes the components, ranking, and overall page optimization for rendering the additional content in the side panes 26. The composer component 22 sends the additional content 28 to the user interface rendering component 10 on the browser for rendering in the side pane 26 and presentation to the user 114.
Rendering component 10 may present additional content in side pane 26 in the vicinity of web page 30. The drawing component 10 can facilitate user interaction with additional content on the side pane 26, such as clicking on a related article or scrolling down or down. Updates to the user profile and session repository based on user interactions may be sent to the runtime server 104 and updated accordingly to support the personalized experience of the user 114.
Rendering component 10 may perform additional processing on the presentation of side pane 26. For example, the drawing component 10 can adjust the size of the side pane 26 (e.g., make the side pane larger or smaller by expanding the width of the side pane, decreasing the width of the side pane, expanding the height of the side pane, and/or decreasing the height of the side pane) based on user input or device display characteristics (e.g., the size of the display). In addition, the drawing component 10 can open the additional content 28 in the new tab based on user input. For example, if the user selects to open a topic of interest icon in the new tab, a web page of the topic of interest may be opened in the new tab.
The environment 100 may have multiple machine learning models running simultaneously. In some implementations, one or more computing devices are used to perform processing for environment 100. The one or more computing devices may include, but are not limited to, server devices, personal computers, mobile devices such as mobile phones, smartphones, PDAs, tablet or notebook computers, and/or non-mobile devices. The features and functionality discussed herein in connection with the various systems may be implemented on one computing device or across multiple computing devices. For example, browser 102, runtime server 104, data source 106, content data store 108, session data store 110, and/or user data store 112 are implemented on the same computing device. Another example includes one or more subcomponents of browser 102, runtime server 104, data source 106, content data store 108, session data store 110, and/or user data store 112 implemented across multiple computing devices. Further, in some implementations, runtime server 104, data source 106, content data store 108, session data store 110, and/or user data store 112 are implemented or processed on different server devices of the same or different cloud computing networks. Furthermore, in some implementations, the features and functionality are implemented or processed on different server devices of the same or different cloud computing networks.
In some implementations, each component of environment 100 communicates with each other using any suitable communication technology. In addition, while the components of environment 100 are shown as separate, any component or sub-component may be combined into fewer components, such as into a single component, or separated into more components, which may serve a particular implementation. In some implementations, the components of environment 100 include hardware, software, or both. For example, components of environment 100 may include one or more instructions stored on a computer-readable storage medium and executable by a processor of one or more computing devices. Computer-executable instructions of one or more computing devices, when executed by one or more processors, may perform one or more methods described herein. In some implementations, the components of environment 100 include hardware, such as a special purpose processing device for performing a particular function or group of functions. In some implementations, the components of environment 100 include a combination of computer-executable instructions and hardware.
Referring now to FIG. 2, an example method 200 for identifying additional content of a web page is illustrated. The actions of method 200 are discussed below with reference to the architecture of FIG. 1.
At 1, method 200 includes opening a web page. The user 114 opens the web page 30 (fig. 1) on the browser 102 by, for example, clicking on the URL of the web page 30.
At 1.1, method 200 includes receiving a click of a button. The user 114 clicks on a toolbar button on the browser 102 for the side pane 26 (fig. 1).
At 2, method 200 includes collecting a URL of a web page. The browser 102 may collect the URL of the web page 30.
At 3, method 200 includes sending a request (e.g., request 24) to runtime server 104. The browser 102 may send a request 24 to the runtime server 104 to present the additional content 28 (fig. 1) in the side pane 26 alongside the web page 30. Request 24 includes the URL of web page 30. The request 24 may be sent in response to the user 114 selecting a button on the browser 102. In addition, the request 24 may be automatically sent in response to the user 114 opening the web page 30.
At 4, method 200 includes classifying the URL. The runtime server 104 receives the request 24 and runs the URL classification to decide whether to trigger the side pane 26 and determine the category and domain of the URL.
At 5, method 200 includes collecting and aggregating content. Based on the web page categories and domains, the runtime server 104 gathers and aggregates content from multiple data providers.
At 6, method 200 includes orchestrating an entire page optimization (WPO). After aggregating all available data, the runtime server 104 performs arbitration and overall page optimization targeting user interactivity to determine the placement of all modules selected for additional content for inclusion in the side pane. The overall page optimization determines the location of the modules and the rank of the modules and determines the placement of the additional content 28 in the side panes 26. The overall page optimization determines the design of the side pane 26 and how the additional content 28 is shown in the side pane 26 to optimize the presentation of the additional content 28 and modules in the side pane 26.
At 7, method 200 includes sending the additional content to the browser. The runtime server 104 sends the organized payload back to the client side (browser 102) for rendering. For active triggers, the side window 26 opens by default (automatically) upon return of data from the runtime server 104.
At 8, method 200 includes sending the user interaction to a runtime server. After loading the first page on the side pane 26, further user interactions with the side pane 26 (e.g., scrolling, clicks in a feed, clicks on a topic, clicks on a knowledge card, etc.) are sent to the same runtime server 104.
At 9, method 200 includes updating the session with the refreshed content. Based on the interaction event, the runtime server 104 updates the session and user profile and obtains additional content 28 with updated information. The refreshed content is sent back to the client side (browser 102) for asynchronous updates on the side pane 26.
If the side pane 26 is open on the browser 102 and the user enters a new URL in the address bar to open a new web page in the current tab, the method 200 may begin at action 2 to collect the new URL and request that the runtime server 104 refresh the side pane 26 with additional content 28 related to the web page of the new URL.
If side pane 26 is open on browser 102 and the user refreshes the current web page to force a reload, method 200 may begin with action 2 to collect URLs to reflect any changes to web page 30. A new request 24 is sent to the runtime server 104 to refresh the side pane 26 for new additional content 28.
For a particular URL, if there is no content from the content store, the side pane 26 does not show any context data for the web page 30 (e.g., if the web page is private), and if a user profile is available, personalized article recommendations are provided in the side pane 26. The side pane 26 only shows trending article recommendations if the user profile is not available.
Referring now to FIG. 3, shown is an example Graphical User Interface (GUI) of browser 102 displaying web page 302 and side pane 304, side pane 304 includes summary module 306 that provides points from web page 302 that summarize the content of web page 302.
The runtime server 104 may determine to display the summary module 306 for the web page 302. The summary module 306 may not fit all web pages. Because of the form of text, the summarization module is not useful for some web pages. For example, summaries for poetry, lyrics, online chat, forum, common questions (FAQ), tax forms, classified ads, shopping pages, religious manager, stock quotes may not be useful to the user 114. Furthermore, the content may not be summarized due to the form of the web page content. For example, summaries of online video games, bank accounts, login pages, hotel reservations, flight schedules, television channels, etc. may not be appropriate.
The digest module 306 may generate a digest based on the original document. The runtime server 104 may select a portion of the web page 302 to generate the summary. In addition, the runtime server 104 may generate a summary of the web page 302 by providing the topic and gist of the document to interpret the content of the web page 302. The topics may be presented in a topic module 308 that identifies topics for the web page 302. The identified topics and key-phrases may be related to the content of the web page 302 and may cover a wide range of categories. The topics and key-phrases may be clickable and the side pane 304 may be refreshed with additional content related to the selected topic or key-phrase. The side pane 304 may include a navigation bar 310 in which the user 114 may return to the previous side pane 304.
Referring now to FIG. 4, shown is an example GUI 400 of the browser 102 displaying a web page 402 and a side pane 408 including a page overview module 404 and a theme module 406. The topic module 406 can present one or more primary topics of the web page 402. The page overview module 404 may provide a summary of the content of the web page 402.
Referring now to FIG. 5, shown is an example GUI 500 of the browser 102 displaying a web page 502 and a side pane 504 including a question and answer module 506. The questions and answers module 506 may enable the user to better understand what the current web page 502 pertains to. The questions selected by the runtime server 104 may include questions that are interesting and less obvious that are not curious to the user; leading to a better understanding of the problem of web page 502; and to address various issues with the entire content of web page 502 (rather than focusing on a subset/portion of the content of web page 502). The questions may be linkable, allowing the user 114 to jump to the actual paragraph in the web page 502 where the answer was extracted.
Referring now to FIG. 6, shown is an example GUI 600 of the browser 102 displaying a web page 602 and a side pane 604 including a knowledge card module 606. The runtime server 104 may trigger a knowledge card for the dominant entity of the web page 602. For example, a knowledge card may be provided for the sports team mentioned in web page 602. If the web page 602 has multiple prominent entities, the runtime server 104 may provide an appropriate experience that highlights all of the prominent entities. The knowledge cards may be dynamic, rotated, and/or updated according to user sessions.
Referring now to FIG. 7, shown is an example GUI 700 of the browser 102 displaying a web page 702 and including a side pane 704 with respect to a source module 706. The information about the publisher or data source of the web page 702 may be provided by the source module 706.
Referring now to FIG. 8, there is illustrated an example GUI 800 of the browser 102 displaying a web page 802 and a side pane 804, the side pane 804 including a recommended content module 806 having content related to the web page 802. The related content may help to motivate the user's interest in the browsing session and keep the user 114 interactive.
The recommended content module 806 may include a portion that provides additional relevant content to the web page 802 that recommends more for you. The runtime server 104 can provide relevant content summaries to present recommended content for the web page 802 on the side pane 804. The recommended content may be related to a topic similar to the content of the web page 802 (e.g., the same topic as the articles on the web page 802). The recommended content may be various content about different topics or subjects. The recommended content may be interesting content with a title or a segment of the content to convey interesting factors of the content. The recommended content may have thumbnail images such that each image is unique in the relevant content feed. The recommended content may be ranked together with fresh content having a higher ranking and placed on top of the relevant content feed. The recommended content may come from authoritative sources and may promote time-insensitive content. The related content feed may have an infinite scroll in which the user 114 may view new content as the user 114 scrolls down the side pane 804. The recommended content may be personalized for the interests of the user and/or based on past interactions of the user with the browser. Recommended content may be added to the page view count on the provided article to entice the user to view the recommended content.
If the user 114 selects one of the articles in the relevant content feed, the browser 102 may open the selected article in the side pane 804. In addition, the browser 102 can open the selected article in the new tab. After the user 114 opens the relevant article, the side window 804 may be refreshed with new page content.
The browser 102 can adjust the width of the side pane 804. In addition, browser 102 can provide one or more overlays on top of side pane 804. For example, if the user clicks on a link to related content, the related content may be presented in an overlay on top of the side pane 804. Browser 102 can also load relevant content in the main window in place of web page 802.
Referring now to FIG. 9, shown is an example GUI 900 of the browser 102 displaying a web page 902 and a side pane 904 including a book module 906. The book module 906 may provide the related books based on the content of the web page 902. By providing relevant books to the content of web page 902, side pane 904 may provide additional contextual information to user 114 and allow faster task completion (e.g., purchasing a book).
Referring now to FIG. 10, illustrated is an example GUI 1000 of the browser 102 displaying a side pane 1002 that includes a shopping module 1004. Side pane 1002 may be displayed next to or near a web page. Shopping module 1004 may include products, coupons, transactions related to the content of the web page. In addition, side pane 1002 may facilitate user 114 purchasing one of the products included in shopping module 1004.
Referring now to FIG. 11, there is illustrated an example GUI 1100 of the browser 102 displaying a side pane 1102 including a travel module 1104. The side pane 1102 may be displayed on one side (e.g., left or right) of the web page. The travel module 1104 can include a map having points of interest related to the content of the web page. The travel module 1104 may also include the weather for the location discussed in the web page. The travel module 1104 can also facilitate purchasing tickets (e.g., flights or trains) to the locations mentioned in the web page.
Referring now to FIG. 12, shown is an example GUI 1200 of the browser 102 displaying a side window 1202 including a food module 1204. Side window 1202 may be presented below or above the web page. Side window 1202 may include recipes or restaurants related to the content of the web page.
Referring now to FIG. 13, illustrated is an example GUI 1300 of an example of a browser 102 displaying a web page 1302 and a side pane 1304 including a cooking module 1306. The web page 1302 may include a recipe for cooking a dish. The side window 1304 may include an overview module with this recipe module regarding information extracted from the content of the web page 1302 for the cooking. The recipe module may include a cooking time (15 minutes), a preparation time (10 minutes), a serving size (6) and a raw material for frying. The machine learning module may automatically extract the recipe information from the content of the web page 1302 to summarize the recipe.
The side window 1304 may also include a cooking module 1306 with an associated recipe. The related recipe may be a single recipe or a summary of related recipes. The relevant recipes may be based on the content of the web page 1302. The relevant recipes may be similar to the recipes in the web page 1302. The summary of related recipes may comprise images of different recipes with dish names. The user 114 may scroll through the summaries of the relevant recipes and may continue to receive additional recipes as the user continues to scroll through the summaries of the relevant recipes on the side window 1304. The user 114 can click on any relevant recipe and load the recipe facts in the side window 1304. The user may also have a web page with related recipes open in side pane 1304. The relevant recipe recommendation may be a recipe from the same publisher as the web page the user is viewing, or may be a recipe from a different publisher than the web page the user is viewing. The machine learning model may determine the most relevant recipes based on, for example, the ingredients, cooking styles mentioned in web page 1302.
Referring now to FIG. 14, an example GUI 1200 of a browser 102 is shown, the browser 102 displaying a web page 1402 and a side pane 1404, the side pane 1404 including video 1406 and images 1408 related to the content of the web page 1402.
Referring now to FIG. 15, illustrated is an example methodology 1500 for automatically identifying additional content of a web page. The actions of method 1500 are discussed below with reference to the architecture of FIG. 1. The acts of method 1500 may be performed by runtime server 104 (fig. 1).
At 1502, method 1500 includes receiving a request for additional content of a web page. The runtime server 104 receives the request 24 for additional content 28 of the web page 30. In some implementations, the request 24 is sent in response to the user 114 selecting a button on the browser 102. In some implementations, the request 24 is automatically sent in response to the user 114 accessing the web page 30 using the browser 102.
At 1504, method 1500 includes categorizing a URL of a web page. Server 104 classifies URLs for web pages 30. In some implementations, classifying the URLs of the web pages 30 is performed by one or more machine learning models. In some implementations, the one or more machine learning models include a transformer machine learning model. In some implementations, the one or more machine learning models include a classification model, a binary model, a regression model, and/or a language model. The machine learning model identifies a domain of the URL of the web page 30 (e.g., travel, sports, cooking, shopping) and determines a classification of the URL of the web page 30 based on the domain.
At 1506, method 1500 includes triggering a side pane experience for the additional content based on the classification of the web page. Server 104 triggers the side pane 26 experience for the additional content 28 based on the classification of web page 30. In some implementations, triggering the side-window-check for additional content is performed by one or more machine learning models. In some implementations, the one or more machine learning models include a transformer machine learning model. In some implementations, the one or more machine learning models include a classification model, a binary model, a regression model, and/or a language model. The machine learning model selects one or more modules for presenting the additional content 28 based on the classification of the URL of the web page 30.
In some implementations, triggering the side pane experience identifies one or more modules to be included in the side pane for presenting additional content. In some implementations, one or more modules are default modules. Server 104 may select different modules for different web page domains. Example modules include, but are not limited to, a summary module, a question and answer module, a recommended content module, a personalized content module, a page overview module, an exploration module, a theme module, a point module, a knowledge card module, a source module, a cooking module, a food module, a travel module, a shopping module, or a book module. In some implementations, the server 104 performs optimization of one or more modules, including ranking the one or more modules for determining an order for presenting the one or more modules in the side pane based on the context of the web page 30. Because of the form and/or content of the web page 30, different modules may be more useful than other modules for the web page 30. For example, if the context of web page 30 is shopping, the summary module has a lower ranking than the exploration module that allows the user to explore more products related to the products on web page 30. Another example includes the summary module and the gist module having a higher ranking than the purchase module if the context of the web page 30 is an article about a virus.
At 1508, method 1500 includes aggregating additional content obtained from one or more data sources for a side pane experience. The server 104 aggregates the additional content 28 obtained from the one or more data sources 106 for a side pane experience. In some implementations, aggregating the additional content 28 for the side pane experience is performed by one or more machine learning models. In some implementations, one or more of the data sources 106 include different data providers for different content. In some implementations, the one or more data sources 106 include a document index or content data store.
At 1510, method 1500 includes sending the additional content to the browser to be presented in a side pane near the web page. The server 104 sends the additional content 28 to the browser 102 to be presented in the side pane 26 beside or near the web page 30. In some implementations, the additional content includes an image or video.
In some implementations, the server 104 performs an entire page optimization to organize the additional content 28 for drawing in the side window 26. The overall page optimization includes applying a ranking to the additional content 28 to determine an order for presenting the additional content 28 in the side window 26. For example, the ranking is based on the relevance of the additional content 28 to the context. In some implementations, relevance is determined by identifying entities or topics in the additional content 28 that match the entities or topics identified in the content of the web page 30. The threshold level is set and if the number of matching entities or topics exceeds the threshold level, the additional content 28 is related to the content of the web page 30. Example threshold levels are four matching entities or topics. If the number of matching entities or topics is below the threshold level, then the additional content 28 is not related to the content of the web page 30. The additional content 28 having matching entities or topics is ranked higher relative to the additional content 28 having a number of matching entities or topics below the threshold level.
In some implementations, the server 104 stores the additional content 28 in the session data store 110 for the user 114. Session data store 110 provides a snapshot of the current user session and identifies additional content 28 that was presented to user 114 in side pane 26 during the current user session.
In some implementations, the server 104 determines whether the request 24 is for a newly accessed web page, and if the request 24 is for a newly accessed web page, the server 104 performs classification of the URL for the web page 30 (1504). If the request 24 comes from a user interaction in the side pane 26 for the web page 30, the server 104 accesses the current user session for the web page 30 from the session data store 110 and sends different additional content 28 to the browser 102 to be presented in the side pane 26 in the vicinity of the web page 30. User interactions include, but are not limited to, clicking on additional content, scrolling up to the top portion of the side pane, or scrolling down to the bottom portion of the side pane. Server 104 stores the user interactions in session data store 110 as part of the current user session for web page 30.
Referring now to FIG. 16, an example method 1600 for presenting additional content of a web page is illustrated. The actions of method 1600 are discussed below with reference to the architecture of FIG. 1. The acts of method 1600 may be performed by browser 102 (fig. 1).
At 1602, method 1600 includes identifying a Uniform Resource Locator (URL) of a web page accessed by a browser. Browser 102 identifies the URL of web page 30 accessed by user 114.
At 1604, method 1600 includes sending a request for additional content. The browser 102 sends a request 24 for additional content 28. Request 24 includes the URL of web page 30. In some implementations, the request 24 for additional content 28 is sent in response to the user 114 selecting a button or icon on the web page that is continuously enabled on the browser 102 while the user 114 is browsing the web site. In some implementations, the request 24 for additional content 28 is automatically sent in response to the browser 102 accessing one or more of the web pages 30 or in response to the URL of the web page 30 matching one or more domains.
At 1606, method 1600 includes receiving additional content. The browser 102 receives the additional content 28. In some implementations, the additional content 28 is based on the context of the web page 30. The context of the web page 30 includes the content of the web page 30.
At 1608, the method 1600 includes presenting the additional content in a side pane next to the web page. The browser 102 presents the additional content 28 in the side pane 26 beside or near the web page 30. In some implementations, the browser 102 resizes the side pane 26 by one or more of expanding the width of the side pane 26, reducing the width of the side pane 26, expanding the height of the side pane 26, or reducing the height of the side pane 26. In some implementations, the browser 102 receives a selection of additional content 28 and opens the selected additional content 28 within the side pane 26 or in a new tab of the browser 102.
As shown in the foregoing discussion, the present disclosure utilizes various terms to describe features and advantages of model evaluation systems. Additional details concerning the meaning of these terms are now provided. For example, as used herein, a "machine learning model" refers to a computer algorithm or model (e.g., classification model, binary model, regression model, language model, object detection model) that can be tuned (e.g., trained) to approximate an unknown function based on training inputs. For example, a machine learning model may refer to a neural network (e.g., convolutional Neural Network (CNN), deep Neural Network (DNN), recurrent Neural Network (RNN)) or other machine learning algorithm or architecture that learns and approximates complex functions and generates an output based on a plurality of inputs provided to the machine learning model. As used herein, a "machine learning system" may refer to one or more machine learning models that cooperatively generate one or more outputs based on corresponding inputs. For example, a machine learning system may refer to any system architecture having a plurality of discrete machine learning components that take into account heterogeneous information or inputs.
Unless specifically described as being implemented in a particular manner, the techniques described herein may be implemented in hardware, software, firmware, or any combination thereof. Any features described as modules, components, or the like may also be implemented together in an integrated logic device or separately as discrete but interoperable logic devices. If implemented in software, the techniques may be realized at least in part by a non-transitory processor-readable storage medium comprising instructions that when executed by at least one processor perform one or more of the methods described herein. The instructions may be organized as a routine, program, object, component, data structure, etc., that may perform a particular task and/or implement a particular data type, and that may be combined or distributed as desired in various implementations.
Computer readable media can be any available media that can be accessed by a general purpose or special purpose computer system. The computer-readable medium storing computer-executable instructions is a non-transitory computer-readable storage medium (device). The computer-readable medium carrying computer-executable instructions is a transmission medium. Thus, by way of example, and not limitation, implementations of the present disclosure may include at least two distinct computer-readable media: a non-transitory computer readable storage medium (device) and a transmission medium.
As used herein, a non-transitory computer-readable storage medium (device) may include RAM, ROM, EEPROM, CD-ROM, a solid state drive ("SSD") (e.g., based on RAM), flash memory, phase change memory ("PCM"), other types of memory, other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store desired program code means in the form of computer-executable instructions or data structures and that can be accessed by a general purpose or special purpose computer.
The steps and/or actions of the methods described herein may be interchanged with one another without departing from the scope of the claims. In other words, unless a specific order of steps or actions is required for proper operation of the method described, the order and/or use of specific steps and/or actions may be modified without departing from the scope of the claims.
The articles "a," "an," and "the" are intended to mean that there are one or more of the elements in the foregoing description. The terms "comprising," "including," and "having" are intended to be inclusive and mean that there may be additional elements other than the listed elements. In addition, it should be understood that references to "one implementation" or "an implementation" of the present disclosure are not intended to be interpreted as excluding the existence of additional implementations that also incorporate the recited features. For example, any element described in connection with an implementation herein may be combined with any element of any other implementation described herein. The numbers, percentages, ratios, or other values recited herein are intended to include the value, and also include other values of the stated value "about" or "approximately," as would be understood by one of ordinary skill in the art encompassed by the implementations of the present disclosure. Accordingly, the stated values should be construed broadly enough to encompass values at least close enough to the stated values to perform the desired function or to achieve the desired result. The stated values include at least the variations expected during suitable manufacturing or production processes, and may include values within 5%, within 1%, within 0.1%, or within 0.01% of the stated values.
Those of ordinary skill in the art should, in light of the present disclosure, appreciate that equivalent constructions do not depart from the spirit and scope of the present disclosure, and that various changes, substitutions, and alterations can be made to the implementations disclosed herein without departing from the spirit and scope of the present disclosure. Equivalent structures including the terms of functional "means plus function" are intended to cover structures described herein as performing the recited function, including both structural equivalents that operate in the same manner and equivalent structures that provide the same function. Applicant expressly intend not to invoke any claim elements plus functions or other functional claims unless the term "element for … …" is present with the associated function. Each addition, deletion, and modification of the implementation that falls within the meaning and scope of the claims is covered by the claims.
Industrial application
The present disclosure relates to systems and methods for identifying additional content of a web page. The system and method creates a side pane that is placed next to or near the web page to supplement the web page that the user is viewing. The side pane provides additional information and/or additional content to the user, and the side pane remains in the front view of the web page. The additional content provides web page insights, such as page summaries, page topics, and/or key entities that help the user consume current web page content and/or recommended context or personalized content (related articles) for further exploration by the user. In some implementations, the additional content is from multiple sources. In some implementations, the additional content is based on the context of the web page that the user is currently browsing. In some implementations, the additional content includes content derived from a web page.
In some implementations, the side panes are reactively opened based on user input. The user may click on a toolbar icon on the browser and, in response to the user clicking on the toolbar icon, a side window with additional content opens. One example includes, in response to a user accessing a web page of a sports team and clicking on a toolbar icon of a side pane, the system identifying an entity card for the sports team as additional content and presenting the side pane with the entity card in proximity to the web page.
In some implementations, the side panes open automatically based on a user accessing a web page. One example includes, in response to a user accessing a web page having an article describing a holding of an evening party, the article including different recipes for the evening party, the system identifying additional content and presenting the additional content in a side pane on the right side of the web page. The additional content includes a recipe mentioned in the article for the evening party. The additional content includes information about the mentioned recipes in the article (e.g. raw materials, cooking time, serving portion).
In some implementations, the side pane has one or more modules that highlight the additional content. The side pane's modules help users consume content, help users explore additional content, and/or help maintain a user interactive browsing experience. In some implementations, the modules included in the side panes vary based on the domain of the web page. For example, news, finance, shopping, weather, maps, sports, shopping, cooking, and travel domains include different modules in the side panes. In some implementations, the modules on the side pane include default modules (e.g., overview module, source module, explorer module) that are included in the web page regardless of the domain of the web page.
In some implementations, the modules of the side pane are generated by a machine learning model that reviews the content of the web page and associates the content of the web page with or with topics that are related to the content mentioned in the web page.
The system and method may be used as a reading assistant and a content discovery engine. The system and method utilizes natural language understanding, machine understanding, and personalized recommendation techniques to correlate content with a user's current tasks and long-term interests. In some implementations, the systems and methods proactively provide relevant information and enable further exploration without requiring the user to be taken off of the current task workflow.
One technical advantage of the system and method is to provide additional content supported by a browser and thus may increase user interactivity with the browser. The system and method improves user efficiency by organizing and arranging additional content into modules. The system and method also reduces the need for a user to navigate to multiple websites and thus reduces user input and/or clicks. The system and method also improves the user's trust because several of the modules allow the user to understand the origin and reputation of the website. In addition, the system and method improves accessibility to websites that otherwise do not have good accessibility options.
The system and method gives the user insight into the currently viewed web documents, enabling deep exploration of the current topic, suggesting new topics, understanding various perspectives, and/or providing other contextual experiences for the user.
(A1) Some implementations include a method for automatically identifying additional content of a web page. The method includes receiving (1502) a request (e.g., request 24) for additional content of a web page (e.g., web page 30), the request including a Uniform Resource Locator (URL) of the web page. The method includes classifying a URL of a web page (1504). The method includes triggering (1506) a side-window experience for the additional content based on the classification of the web page. The method includes aggregating (1508) additional content (e.g., additional content 28) obtained from one or more data sources (e.g., data source 106) for a side-window experiment, the additional content being obtained based on a context of the web page. The method includes sending (1510) additional content to a browser (e.g., browser 102) to be presented in a side pane (e.g., side pane 26) in proximity to a web page (e.g., web page 30).
(A2) In some implementations of the method of A1, the request is sent in response to a user selecting a button on the browser.
(A3) In some implementations of the methods of A1 or A2, the request is sent automatically in response to a user accessing the web page using a browser.
(A4) In some implementations of the method of any of A1-A3, classifying the URLs of the web pages is performed by one or more machine learning models by identifying domains of the URLs of the web pages and determining the classification of the URLs of the web pages based on the domains; and the one or more machine learning models include a classification model, a transformer model, a binary model, a regression model, or a language model.
(A5) In some implementations of the method of any of A1-A4, triggering the side pane experience identifies one or more modules to be included in the side pane for presenting additional content, and selecting different modules for different web page domains.
(A6) In some implementations of the method of any of A1-A5, one or more modules are default modules.
(A7) In some implementations of the method of any of A1-A6, the one or more modules include a summary module, a question and answer module, a recommended content module, a personalized content module, a page overview module, an exploration module, a topic module, a point module, a knowledge card module, a source module, a cooking module, a food module, a travel module, a shopping module, or a book module.
(A8) In some implementations, the method of any of A1-A7 includes performing an optimization of the one or more modules, the optimization including ranking the one or more modules for determining an order for presenting the one or more modules in the side pane based on a context of the web page.
(A9) In some implementations of the method of any of A1-A8, the additional content includes an image or video.
(A10) In some implementations, the method of any of A1-A9 includes performing an entire page optimization to organize additional content for drawing in the side panes.
(A11) In some implementations of the method of any of A1-a10, the overall page optimization includes applying a ranking to the additional content based on relevance of the additional content to the context of the web page to determine an order for presenting the additional content in the side window.
(A12) In some implementations of the method of any of A1-a11, determining relevance by identifying entities in the additional content that match the identified entities in the web page and comparing the matching entities to a threshold level; if the matching entity exceeds the threshold level, determining that the additional content is relevant to the context of the web page and providing a higher ranking to the additional content; and if the matching entity is below the threshold level, determining that the additional content is not relevant to the context of the web page and providing a lower ranking to the additional content.
(A13) In some implementations, the method of any of A1-a12 includes storing the additional content in a session data store for the user, and the session data store provides a snapshot of a current user session and identifies the additional content presented to the user in a side pane during the current user session; determining whether the request is for a newly accessed web page; if the request is for a newly accessed web page, performing classification of the URL of the web page; and if the request is from a user interaction in a side pane for the web page, accessing a current user session for the web page from a session data store and sending different additional content to a browser to be presented in the side pane in the vicinity of the web page.
(A14) In some implementations, the method of any of A1-a13 includes storing the user interaction in the session data store as part of a current user session for the web page, and the user interaction includes one or more of clicking on additional content, scrolling up to a top portion of the side pane, or scrolling down to a bottom portion of the side pane.
(A15) In some implementations of the method of any of A1-a14, the one or more data sources include different data providers for different content, and the one or more data sources include a document index or content data store.
(A16) In some implementations of the method of any of A1-a15, triggering the side pane experience for the additional content and aggregating the additional content for the side pane experience is performed by one or more machine learning models selecting a plurality of modules for presenting the additional content based on the identified domains for the web page.
(B1) Some implementations include a method for presenting additional content of a web page. The method includes identifying (1602) a Uniform Resource Locator (URL) of a web page (e.g., web page 30) accessed by a browser (e.g., browser 102). The method includes sending (1604) a request for additional content (e.g., request 24), where the request includes a URL of a web page. The method includes receiving (1606) additional content (e.g., additional content 28) based on a web page context of the web page. The method includes presenting (1608) additional content in a side pane (e.g., side pane 26) beside the web page.
(B2) In some implementations of the method of B1, the request for additional content is sent in response to a user selecting a button or icon on the browser that is continuously enabled on the browser while the user is browsing the website.
(B3) In some implementations of the methods of B1 or B2, sending the request for additional content occurs automatically in response to the browser accessing one or more of the web pages or in response to the URL of the web page matching one or more domains.
(B4) In some implementations, the method of any of B1-B3 includes adjusting the size of the side pane by one or more of expanding the width of the side pane, reducing the width of the side pane, expanding the height of the side pane, or reducing the height of the side pane.
(B5) In some implementations, the method of any of B1-B4 includes receiving a selection of additional content; and opens the selected additional content in the side pane or in a new tab of the browser.
Some implementations include a system (e.g., environment 100). The system includes one or more processors; a memory in electronic communication with the one or more processors; and instructions stored in memory that are executable by the one or more processors to perform any of the methods described herein (e.g., A1-a15, B1-B5).
Some implementations include instructions stored on a computer-readable storage medium that are executable by one or more processors to perform any of the methods described herein (e.g., A1-A16, B1-B5).
Some implementations include a browser (e.g., browser 102) executable by one or more processors to perform any of the methods described herein (e.g., A1-A16, B1-B5).
Some implementations include a browser (e.g., browser 104) executable by one or more processors to perform any of the methods described herein (e.g., A1-A16, B1-B5).
The present disclosure may be embodied in other specific forms without departing from its spirit or characteristics. The described implementations should be regarded as illustrative rather than restrictive. The scope of the disclosure is, therefore, indicated by the appended claims rather than by the foregoing description. Changes which come within the meaning and range of equivalency of the claims are to be embraced within their scope.
Claims (15)
1. A method for automatically identifying additional content of a web page, comprising:
receiving a request for additional content of the web page, wherein the request includes a Uniform Resource Locator (URL) of the web page;
classifying the URLs of the web pages;
triggering a side pane experience for the additional content based on the classification of the web page;
aggregating additional content for the side pane experience obtained from one or more data sources, wherein the additional content is obtained based on a context of the web page; and
The additional content is sent to a browser to be presented in a side pane near the web page.
2. The method of claim 1, wherein the request is sent in response to a user selecting a button on the browser, or
Wherein the request is automatically sent in response to a user accessing the web page using the browser.
3. The method of claim 1, wherein classifying the URLs of the web pages is performed by one or more machine learning models by identifying a domain of the URLs of the web pages and determining a classification of the URLs of the web pages based on the domain, and
Wherein the one or more machine learning models include a classification model, a transformer model, a binary model, a regression model, or a language model.
4. The method of claim 1, wherein triggering the side pane experience identifies one or more modules to be included in the side pane for presenting the additional content,
Wherein different modules are selected for different web page domains and the one or more modules include a summary module, a question and answer module, a recommended content module, a personalized content module, a page overview module, a exploration module, a theme module, a point module, a knowledge card module, a source module, a cooking module, a food module, a travel module, a shopping module, or a book module.
5. The method of claim 4, wherein the one or more modules are default modules.
6. The method of claim 4, further comprising:
An optimization of the one or more modules is performed, wherein the optimization includes ranking the one or more modules for determining an order for presenting the one or more modules in the side pane based on the context of the web page.
7. The method of claim 1, wherein the additional content comprises an image or video.
8. The method of claim 1, further comprising:
Performing an overall page optimization to organize the additional content for rendering in the side pane, wherein the overall page optimization includes applying a ranking to the additional content to determine an order for rendering the additional content in the side pane based on a relevance of the additional content to the context of the web page, wherein the relevance is determined by identifying entities in the additional content that match entities identified in the web page and comparing the matching entities to a threshold level;
Determining that the additional content is relevant to the context of the web page and providing a higher ranking to the additional content if the matching entity exceeds the threshold level; and
If the matching entity is below the threshold level, determining that the additional content is not relevant to the context of the web page and providing a lower ranking to the additional content.
9. The method of claim 1, further comprising:
Storing the additional content in a session data store for a user, wherein the session data store provides a snapshot of a current user session and identifies the additional content presented to the user in the side pane during the current user session;
determining whether the request is for a newly accessed web page;
If the request is for a newly accessed web page, performing the classification of the URL of the web page; and
If the request is from a user interaction in the side pane for the web page, accessing the current user session for the web page from the session data store, sending different additional content to the browser to be presented in the side pane in the vicinity of the web page, and storing the user interaction in the session data store as part of the current user session for the web page, wherein the user interaction includes one or more of clicking on the additional content, scrolling up to a top portion of the side pane, or scrolling down to a bottom portion of the side pane.
10. The method of claim 1, wherein the one or more data sources comprise different data providers for different content and the one or more data sources comprise a document index or a content data store.
11. The method of claim 1, wherein triggering the side pane experience for the additional content and aggregating the additional content for the side pane experience are performed by one or more machine learning models selecting a plurality of modules for presenting the additional content based on the identified domains for the web page.
12. A system, comprising:
A memory storing data and instructions; and
At least one processor operable to communicate with the memory, wherein the at least one processor is operable to:
Receiving a request for additional content of a web page, wherein the request includes a Uniform Resource Locator (URL) of the web page;
classifying the URLs of the web pages;
triggering a side pane experience for the additional content based on the classification of the web page;
aggregating additional content for the side pane experience obtained from one or more data sources, wherein the additional content is obtained based on a context of the web page; and
The additional content is sent to a browser to be presented in a side pane near the web page.
13. The system of claim 12, wherein the at least one processor is further operable to:
categorizing the URLs of the web pages by identifying a domain of the URLs of the web pages and determining a categorization of the URLs of the web pages based on the domain; and
The side pane experience is triggered by identifying one or more modules to be included in the side pane for presenting the additional content based on the identified domains for the web page, wherein different modules are selected for different domains.
14. The system of claim 12, wherein the at least one processor is further operable to:
Performing an overall page optimization to organize the additional content for rendering in the side pane, wherein the overall page optimization includes applying a ranking to the additional content based on a relevance of the additional content to the context of the web page to determine an order for rendering the additional content in the side pane.
15. A computer-readable medium storing instructions executable by a computer device, comprising:
At least one instruction for causing the computer device to receive a request for additional content of a web page, wherein the request includes a Uniform Resource Locator (URL) of the web page;
At least one instruction for causing the computer device to categorize the URL of the web page;
At least one instruction for causing the computer device to trigger a side pane experience for the additional content based on the classification of the web page;
At least one instruction for causing the computer device to aggregate additional content for the side pane experience obtained from one or more data sources, wherein the additional content is obtained based on a context of the web page; and
At least one instruction for causing the computer device to send the additional content to a browser to be presented in a side pane in proximity to the web page.
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US63/270,459 | 2021-10-21 | ||
| US17/556,989 US11822612B2 (en) | 2021-10-21 | 2021-12-20 | Automatic identification of additional content for webpages |
| US17/556,989 | 2021-12-20 | ||
| PCT/US2022/044472 WO2023069219A1 (en) | 2021-10-21 | 2022-09-23 | Automatic identification of additional content for webpages |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118103831A true CN118103831A (en) | 2024-05-28 |
Family
ID=91149496
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202280068964.2A Pending CN118103831A (en) | 2021-10-21 | 2022-09-23 | Automatic identification of additional content of a web page |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118103831A (en) |
-
2022
- 2022-09-23 CN CN202280068964.2A patent/CN118103831A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11263217B2 (en) | Method of and system for determining user-specific proportions of content for recommendation | |
| RU2725659C2 (en) | Method and system for evaluating data on user-element interactions | |
| US9374396B2 (en) | Recommended content for an endorsement user interface | |
| US7885986B2 (en) | Enhanced browsing experience in social bookmarking based on self tags | |
| US8495058B2 (en) | Filtering social search results | |
| Aamir et al. | Recommendation system: state of the art approach | |
| US20150317398A1 (en) | Presenting non-suggested content items to a user of a social network account | |
| US20150310392A1 (en) | Job recommendation engine using a browsing history | |
| US20090319512A1 (en) | Aggregator, filter, and delivery system for online content | |
| US20090281988A1 (en) | Systems and Methods to Provide Search Based on Social Graphs and Affinity Groups | |
| US20130060858A1 (en) | Additional Systems and Methods for Curating Content | |
| US20110093520A1 (en) | Automatically identifying and summarizing content published by key influencers | |
| CA2788802A1 (en) | Method and system for providing actionable relevant recommendations | |
| US20180060822A1 (en) | Online and offline systems for job applicant assessment | |
| RU2714594C1 (en) | Method and system for determining parameter relevance for content items | |
| US20140156623A1 (en) | Generating and displaying tasks | |
| Akhil et al. | A survey of recommender system types and its classification. | |
| Nanas et al. | Personalised news and scientific literature aggregation | |
| US20100125809A1 (en) | Facilitating Display Of An Interactive And Dynamic Cloud With Advertising And Domain Features | |
| US11514123B2 (en) | Information retrieval system, method and computer program product | |
| US11822612B2 (en) | Automatic identification of additional content for webpages | |
| WO2018140883A1 (en) | Systems and methods for enhanced online research | |
| Bogers | Recommender systems for social bookmarking | |
| US20180137587A1 (en) | Contextual personalized list of recommended courses | |
| Chidambarathanu et al. | Predicting user preferences on changing trends and innovations using SVM based sentiment analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication |