Browsing history management
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- MICROSOFT TECHNOLOGY LICENSING LLC
- Filing Date
- 2025-01-31
- Publication Date
- 2026-08-06
Smart Images

Figure CN2025075618_06082026_PF_FP_ABST
Abstract
Description
BROWSING HISTORY MANAGEMENTBACKGROUND
[0001] With the development of computer technologies and network technologies, users are increasingly acquiring information through networks. Applications such as browsers may facilitate access to various websites via the networks. For example, users may open and view webpages including various content through browsers. Browsers are usually designed to record browsing histories, which allow users to find previously browsed webpages.SUMMARY
[0002] This Summary is provided to introduce a selection of concepts that are further described below in the Detailed Description. It is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.
[0003] Embodiments of the present disclosure propose methods and apparatuses for browsing history management. Raw data of browsing history may be retrieved. Reference data may be obtained. The reference data may include at least one portion of the raw data. The at least one portion of the raw data may include a record of webpages being browsed within a predetermined period of time. Content data of each webpage may be obtained through a language model. Browsing intention for each webpage may be extracted from the content data through the language model. A set of tasks may be generated through the language model. Each task may include at least two pages. The at least two pages may be a subset of the webpages, and associated with the same browsing intention.
[0004] It should be noted that the above one or more aspects comprise the features hereinafter fully described and particularly pointed out in the claims. The following description and the drawings set forth in detail certain illustrative features of the one or more aspects. These features are only indicative of the various ways in which the principles of various aspects may be employed, and this disclosure is intended to include all such aspects and their equivalents.BRIEF DESCRIPTION OF THE DRAWINGS
[0005] The disclosed aspects will hereinafter be described in connection with the appended drawings that are provided to illustrate and not to limit the disclosed aspects.
[0006] FIG. 1 illustrates an exemplary process for browsing history management according to an embodiment.
[0007] FIG. 2 illustrates an exemplary process for task creation and action suggestion determination according to an embodiment.
[0008] FIG. 3A illustrates an exemplary process for generating visual representations corresponding to a set of tasks according to an embodiment.
[0009] FIG. 3B illustrates an exemplary process for generating visual representations corresponding to a set of action suggestions according to an embodiment.
[0010] FIG. 4 illustrates an exemplary process for generating and loading visual representations according to an embodiment.
[0011] FIG. 5 illustrates an exemplary process for task searching and visual representation loading according to an embodiment.
[0012] FIG. 6 illustrates an exemplary process for page resumption according to an embodiment.
[0013] FIG. 7 illustrates an exemplary process for page quick launch according to an embodiment.
[0014] FIG. 8 illustrates exemplary user interfaces according to an embodiment.
[0015] FIG. 9 illustrates exemplary user interfaces according to an embodiment.
[0016] FIG. 10 illustrates an exemplary user interface according to an embodiment.
[0017] FIG. 11 illustrates an exemplary user interface according to an embodiment.
[0018] FIG. 12 illustrates exemplary user interfaces according to an embodiment.
[0019] FIG. 13 illustrates a flowchart of an exemplary method for browsing history management according to an embodiment.
[0020] FIG. 14 illustrates an exemplary apparatus for browsing history management according to an embodiment.
[0021] FIG. 15 illustrates an exemplary apparatus for browsing history management according to an embodiment.DETAILED DESCRIPTION
[0022] The present disclosure will now be discussed with reference to several example implementations. It is to be understood that these implementations are discussed only for enabling those skilled in the art to better understand and thus implement the embodiments of the present disclosure, rather than suggesting any limitations on the scope of the present disclosure.
[0023] Existing browsers are generally designed to record browsing histories. A browsing history is typically a chronological list of webpages that a user has previously browsed, which aims to help the user find one or more previously browsed webpages. However, such a browsing history may have limited usefulness. For example, when the user is looking for a previously browsed webpage, the user will have to search through a list of webpages in a browsing history and try to narrow the search by relying on some associated information such as browsing time or webpage title, if the user remembers. These operations would be cumbersome and time-consuming and may be particularly challenging if the browsing history includes a large number of previously browsed webpages. In some cases, the browsing history may even lose its effectiveness, for example, if the user wishes to recall or continue a previous activity that occurred across multiple different webpages browsed over a long period of time, the browsing history cannot provide effective help to the user.
[0024] Embodiments of the present disclosure propose browsing history management. According to the embodiments of the present disclosure, a browsing history may be organized based on tasks. A browsing history may include a list of webpages that were browsed via a browser. A task may refer to a collection of webpages that is browsed under a specific intention or purpose. For example, webpages having the same browsing intention may be organized into a task. Hence, webpages included in the browsing history may be organized into different tasks based on respective browsing intentions of these webpages. Herein, webpages organized into a task may also be referred to as pages, so as to literally distinguish from webpages included in the browsing history. As such, a task may include pages associated with the same browsing intention, and these pages may be a subset of the webpages included in the browsing history. Herein, browsing history management may cover various task-related operations or processing to browsing history, e.g., task creating, task-based action suggestion determination, task searching, task-based page resumption, task-based page quick launch, etc. Compared with an existing mechanism that simply establishes a browsing history with a list of webpages, the technical effects of the embodiments of the present disclosure at least include intuitively and efficiently managing a browsing history in the form of tasks, thereby greatly improving the usefulness of such task-based browsing history. In an aspect, with a task-based browsing history, a user can easily find a previously browsed page by means of a task including this page, or can effortlessly recall multiple related pages based on a task. In an aspect, the user can easily reopen all pages included in a task together by interacting with the task on a user interface of the browser, without the need of manually reopening these pages one by one. Hence, the technical effects of the embodiments of the present disclosure further include effectively simplifying users’ navigation operation through the browsing history and significantly enhancing user browsing experience.
[0025] Herein, a browsing intention may refer to a purpose or goal that a user desires to achieve through webpage browsing. A browsing intention may reflect both an activity the user wishes to carry out and objects involved in the activity. As an example, when the user is browsing a shopping webpage with an item query “keyboard” , a browsing intention may be “shopping for a keyboard” , wherein “shopping” is an activity that the user wishes to carry out, and “keyboard” is an object involved in the activity of “shopping” . As another example, when the user is browsing a webpage of introduction to scenic spots in City Y, a browsing intention may be “planning a trip to City Y” , wherein “planning a trip” is an activity that the user wishes to carry out, and “City Y” is an object involved in the activity of “planning a trip” .
[0026] In some implementations, raw data of a browsing history may be retrieved. The raw data of the browsing history may refer to original data or a log for recording that webpages were browsed via a browser. For example, the raw data may include data associated with webpages that have been browsed, and / or data associated with interaction operations to these webpages. Then reference data may be obtained based on the raw data. The reference data may refer to data required for, e.g., task creation, etc. For example, the reference data may include at least one portion, within a predetermined period of time, of the raw data. The at least one portion of the raw data may include a record of webpages being browsed within the predetermined period of time. The record may include webpage information and interaction information of the webpages browsed within the predetermined period of time. Then, content data of each webpage of the webpages browsed within the predetermined period of time may be obtained. Content data of a webpage may refer to data representing content contained in the webpage. As an example, the content data may include text data, image data, multimedia data, etc. Based on content data of each webpage, a browsing intention for the webpage may be determined. For example, the browsing intention may be extracted from the content data. Based on respective browsing intentions of the webpages browsed within the predetermined period of time, a set of tasks may be determined. For example, a set of tasks may be generated, wherein each task may include at least two pages, and the at least two pages may be a subset of the webpages browsed within the predetermined period of time and associated with the same browsing intention.
[0027] In some implementations, at least one operation included in the process described above may be performed through a language model. Different operations may be performed through the same language model or through different language models, and the embodiments of the present disclosure are not limited in this regard. Herein, the language model may refer to a deep learning model that can understand meaning of natural language, generate natural language text, or perform other functions related to natural language. The input and / or output of the language model may have various modalities such as text, audio, image, video, etc. As an example, the language model may be Generative Pre-trained Transformer (GPT) such as GPT-4 and its evolution. The embodiments of the present disclosure are not limited to any specific type of language model, and the term “language model” may cover any suitable language models. Since the language model has various powerful capabilities such as a semantic understanding capability, a logical inference capability, etc., using the language model to perform an operation included in the above process can efficiently provide more accurate and reliable results
[0028] In an implementation, the content data of each webpage may be obtained through a language model. The technical effect of such an implementation is to improve efficiency of obtaining the content data.
[0029] In an implementation, the browsing intention for each webpage may be extracted from content data of the webpage, through a language model. For example, semantics of the webpage may be deeply understood by the language model, and thus, the technical effect of such an implementation lies in improving accuracy and reliability of browsing intentions.
[0030] In an implementation, the set of tasks may be generated through a language model, and the technical effect of such an implementation lies in making the tasks more reasonable and better align with user needs.
[0031] In an aspect, at least one quick launch option corresponding to at least one task may be provided on a user interface of a browser. The technical effect of providing quick launch options lies in enabling a user to reopen all pages included in a target task conveniently without the need of complex interactive operations.
[0032] The embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.
[0033] FIG. 1 illustrates an exemplary process 100 for browsing history management according to an embodiment. The process 100 may be implemented in a browser. Herein, the browser may refer to various types of application that are used for assessing web-based resources and can present webpage contents. The browser may be installed in any suitable device, such as a desktop computer, a laptop computer, a tablet computer, a mobile phone, a wearable device, or any other browser-enabled device. In the process 100, tasks or both tasks and corresponding action suggestions may be determined.
[0034] As shown in FIG. 1, the process 100 may include reference data obtaining at 120. The reference data obtaining at 120 may include obtaining reference data 102 based on raw data 101 of a browsing history. The raw data 101 may be maintained by the browser, thus, the raw data 101 may be retrieved from the browser. The raw data 101 may refer to data for recording multiple webpages that a user has browsed and the user’s interactions with the multiple webpages. For example, the raw data 101 may include webpage information and interaction information of the multiple webpages. The webpage information may include various types of information about webpages, such as, webpage titles, webpage links, etc., of the multiple webpages. A webpage link may include a uniform resource locator (URL) , etc. The interaction information may include various types of information about interactions to webpages, such as, webpage browsing time points, webpage browsing durations, etc., of the multiple webpages. A webpage browsing time point of a webpage may indicate when the webpage was browsed. A webpage browsing duration of a webpage may refer to a length of time during which the webpage was browsed, that is, a length of time that the user spent on the webpage. The reference data 102 may include at least one portion, within a predetermined period of time, of the raw data 101. As such, the reference data 102 may include a portion or all of the raw data 101. For example, if the raw data 101 itself is a record within the predetermined period of time, the reference data 102 may include the whole raw data 101. For another example, if the raw data 101 itself is a record within a longer period than the predetermined period of time, the raw data 101 may be divided by time periods, and a portion within the predetermined period of time may be included in the reference data 102. Accordingly, the at least one portion of the raw data 101 may include a record of webpages being browsed within the predetermined period of time. As described above, the raw data 101 may include webpage information and interaction information of the multiple webpages, therefore, the record may include webpage information and interaction information of the webpage browsed within the predetermined period of time.
[0035] In an implementation, the reference data 102 may include additional data required for task creation, depending on actual application scenarios, technical needs, etc. As an example, the reference data 102 may include at least a portion of content data of one or more webpages of the webpages browsed within the predetermined period of time. As another example, the reference data 102 may include respective webpage indexes of the webpages browsed within the predetermined period of time. For instance, webpage indexes may be assigned to the webpages based on, e.g., an ascending order of browsing time points of the webpages.
[0036] Herein, the predetermined period may have any duration, which may be set according to various factors, such as actual application scenarios, technical needs, a constraint on input data volume of a language model (e.g., the language model 180) , etc. As an example, the predetermined period may be 30 minutes, several hours, one day, etc.
[0037] The process 100 may further include task creation at 140. The task creation at 140 may include creating a set of tasks 103 based on the reference data 102. In an implementation, the task creation at 140 may performed through a language model 180. Exemplary implementations of the task creation will be described in detail below with reference to FIG. 2.
[0038] Optionally, after creating the set of tasks 103, the process 100 may further include action suggestion determination at 160. The action suggestion determination at 160 may include determining a set of action suggestions 104 for a task. An action suggestion may indicate an action that can be performed for the task. Such an action suggestion would be helpful for a user to conveniently take a corresponding action, thereby further enhancing user browsing experience. In an implementation, a set of action suggestions may be determined for each task of the set of tasks 103. In an implementation, a set of action suggestions may not be determined for each task. Whether an action suggestion is determined for a task may depend on actual application scenarios, technical needs, etc. Although not shown in FIG. 1, the action suggestion determination may also be performed through a language model, such as the language model 180. Exemplary implementations of the action suggestion determination will be described in detail below with reference to FIG. 2.
[0039] It should be appreciated that the process 100 is only exemplary, and according to actual application requirements and designs, the process 100 may comprise more or less modules and operations, or various changes may be made to the process 100.
[0040] FIG. 2 illustrates an exemplary process 200 for task creation and action suggestion determination according to an embodiment. In the process 200, the steps 242, 244 and 246 may correspond to the step 140 shown in FIG. 1, and the step 260 may correspond to the step 160 shown in FIG. 1. Through the steps 242, 244 and 246, a set of tasks may be created. Through the step 260, a set of action suggestions for a task may be determined.
[0041] As shown in FIG. 2, the process 200 may include content data obtaining at 242. The content data obtaining at 242 may include obtaining content data 202 of each webpage identified by reference data 201. The reference data 201 may correspond to the reference data 101 shown in FIG. 1. As described above with regard to the reference data 101, the reference data 201 may include a record of webpages being browsed within a predetermined period of time. Thus, at 242, for each webpage of the webpages browsed within a predetermined period of time, the content data 202 may be obtained. In an implementation, for each webpage, a webpage link may be read from the reference data 201, the webpage may be accessed by utilizing the webpage link, and then the content data of the webpage may be obtained. In another implementation, as described above with regard to the reference data 101, the reference data 201 may include additional data such as at least a portion of content data of a webpage. In this case, the content data of the webpage may be directly obtained from the reference data 201.
[0042] The content data 202 of each webpage may refer to data representing content that may be contained in the webpage and presented on the webpage. For example, the content data 202 may include text data, image data, multimedia data (e.g., video data, audio data) , etc.
[0043] In an implementation, the content data obtaining at 242 may be performed through a language model, such as the language model 180. In this case, a prompt associated with content data obtaining may be provided to the language model. The prompt and the reference data 201 may be separately provided to the language model, or may be integrated together as input data to be provided to the language model. The prompt may include a set of instructions on how to obtain content data. As an example, the prompt may include a set of rules for obtaining content data, several examples of content data, etc. The set of rules for obtaining content data may specify obtaining content data by utilizing a webpage link included in reference data, directly obtaining content data from reference data, etc. The language model may conduct learning for the examples of the content data in the prompt, understand the set of rules in the prompt, and obtain content data based on the reference data. By utilizing the language model, efficiency of obtaining content data can be improved.
[0044] The process 200 may further include browsing intention extraction at 244. The browsing intention extraction at 244 may include, for each webpage of the webpages browsed within the predetermined period of time, determining a browsing intention 203 based on the content data 202. For example, the browsing intention 203 may be extracted from the content data 202. As an example, if content data of a webpage indicates content related to artificial intelligence (AI) technologies, a browsing intention for the webpage may be extracted as “learning AI technologies” . As another example, if content data of a webpage indicates content related to various keyboards and their prices, a browsing intention for the webpage may be extracted as “shopping for a keyboard” .
[0045] In an implementation, the browsing intention 203 may be extracted from the content data 202 through a language model, such as the language model 180. In this case, a prompt associated with browsing intention extraction may be provided to the language model. The prompt and the content data 202 may be separately provided to the language model, or may be integrated together as input data to be provided to the language model. The prompt may include a set of instructions on how to extract browsing intentions. As an example, the prompt may include a set of rules for extracting browsing intentions, several examples of browsing intentions, etc. The set of rules for extracting browsing intentions may specify extracting browsing intentions based on semantics of content data, making browsing intentions include activities and objects involved in the activities, etc. Based on the content data of each webpage and the prompt, the language model may extract a browsing intention from the content data. For example, the language model may conduct learning for the examples in the prompt, understand the set of rules in the prompt, and extract a browsing intention from the content data by utilizing its various capabilities such as a semantic understanding capability, a logical inference capability, etc. In this way, the browsing intention can be accurately determined.
[0046] The process 200 may further include task generation at 246. The task generation at 246 may include generating a set of tasks 204 based on respective browsing intentions for the webpages browsed within the predetermined period of time. For example, at least two webpages having the same browsing intention may be organized into a task. The set of tasks 204 may correspond to the set of tasks 103 shown in FIG. 1. As described above, the term “page” may refer to a webpage included in a task. Each task may include at least two pages associated with the same browsing intention, and the at least two pages may be a subset of the webpages browsed within the predetermined period of time. Moreover, if there are multiple tasks, pages included in one task may be different from pages included in any other task.
[0047] In an implementation, the task generation at 246 may be performed through a language model, such as the language model 180. In this case, a prompt associated with task generation may be provided to the language model. The prompt and the browsing intentions for the webpages browsed within the predetermined period of time may be separately provided to the language model, or may be integrated together as input data to be provided to the language model. The prompt may include a set of instructions on how to organize webpages into tasks. For example, the prompt may include a set of rules for organizing webpages into tasks, several examples of organizing webpages into tasks, etc. The set of rules for organizing webpages into tasks may specify organizing webpages into tasks based on browsing intentions, skipping login webpages when organizing webpages, etc. The language model may generate the set of tasks 204 based on the browsing intentions and the prompt. For example, the language model may conduct learning for the examples in the prompt, understand the sets of rules in the prompt, and organize the webpage into the set of tasks 204, by utilizing its various capabilities such as a semantic understanding capability, a logical inference capability, etc. In this way, task can be accurately and efficiently generated.
[0048] In an implementation, the language model may use some indicators to indicate pages included in each task, in order to reduce data amount used in the language model. For example, the indicators may include page titles, page indexes (e.g., if the reference data 201 includes webpage indexes) , etc.
[0049] In an implementation, each task may be represented by a set of characteristics. For example, for each task, a set of characteristics may include a task name, a task category, page quantity of pages included in the task, a page title of each page, a page link of each page, a page browsing time point, a page browsing duration, etc. The task name may indicate, e.g., a browsing intention of the task. The browsing intention of the task may be the same browsing intention that the pages included in the task are associated with. The task category may refer to a category or classification the task belongs to. As an example, the task category may include entertainment, shopping, learning, healthcare, travel, etc. The page browsing time point may be, e.g., a time point at which the last page among the pages included in the task was browsed, a time point at which the first page among the pages was browsed, etc. The page browsing duration may be a length of time that a user stayed on a page (e.g., the first page, the last page) of the pages, an average length of time that the user stayed on the pages, etc.
[0050] In the implementation of representing a task with a set of characteristics, the task generation at 246 may include organizing the webpages into the set of tasks and determining a set of characteristics of each task. If a language model is utilized to perform the step 246, the prompt associated with task generation may also include a set of rules for generating task names, a set of rules for determining task categories, etc. The set of rules for generating task names may specify a length of a task name, starting with a verb, avoiding vogue words, etc. The set of rules for determining task categories may specify selecting a task category from a given list of task categories. For example, the list of task categories may include predefined categories, such as, entertainment, shopping, learning, healthcare, travel, etc.
[0051] Optionally, the process 200 may include action suggestion determination at 260. The action suggestion determination at 260 may include determining a set of action suggestions 205 for a task. The set of action suggestions 205 may correspond to the set of action suggestions 104 shown in FIG. 1. For a task, the set of action suggestions 205 may be determined based on a browsing intention of the task and / or page content data of the task. As described above, a task may include at least two pages associated with the same browsing intention, as such, each task may have its browsing intention. The page content data of the task may include content data of one or more pages included in the task. In an implementation, the set of action suggestions 205 may be determined based on the browsing intention of the task. For example, if the browsing intention is “shopping for a keyboard” , an action suggestion may be determined as suggesting an action for shopping for a keyboard. In another implementation, the set of action suggestions 205 may be determined based on the browsing intention of the task and the page content data of the task. For example, if the browsing intention is “planning a trip to City Y” and the page content data includes image data indicating a distance between City Z and City Y, an action suggestion may be determined as suggesting an action for booking a flight from City Z to City Y. In another implementation, the set of action suggestions 205 may be determined based on the page content data. For example, if the page content data includes text data describing development of AI technologies, an action suggestion may be determined as suggesting an action for searching for best tools of AI technologies.
[0052] In an implementation, the action suggestion determination at 260 may include generating the set of action suggestions 205 based on at least one action type and at least one corresponding action parameter. An action type may refer to a type or class to which an action belongs. An action parameter may refer to a variable or an object involved when an action is taken. Each action type may correspond to at least one action parameter. The at least one action type and the at least one corresponding action parameter may be extracted from the browsing intention and / or the page content data of the task. For example, if the browsing intention is “shopping for a keyboard” , an action type may be “shopping” , and an action parameter may include an object “keyboard” . For another example, if the page content data includes text data describing development of AI technologies, an action type may be “searching” , and an action parameter may include a search query “AI technologies” . For another example, if the browsing intention is “planning a trip to City Y” and the page content data includes image data indicating a distance between City Z and City Y, an action type may be “booking a flight” , and action parameters may include a departure city “City Z” and a destination city “City Y” .
[0053] In an implementation, the action suggestion determination at 260 may be performed through a language model, such as the language model 180. By utilizing various powerful capabilities of the language model, action suggestions can be determined that better align with user needs.
[0054] For example, the at least one action type and the at least one corresponding action parameter may be extracted from the browsing intention and / or the page content data of the task, through a language model, such as the language model 180. In this case, a prompt associated with action suggestion determination may be provided to the language model. The prompt and at least one of the browsing intention and page content data of the task may be separately provided to the language model, or may be integrated together as input data to be provided to the language model. The prompt may include a set of instructions on how to extract action types and corresponding action parameters. For example, the prompt may include a set of rules for extracting action types, a set of rules for extracting action parameters, etc. The set of rules for extracting action types may specify extracting action types from a browsing intention and / or page content data of a task, selecting an action type from a given list of action types based on a browsing intention and / or page content data of a task, etc. As an example, the list of action types may include predefined action types, such as, searching, booking a hotel, booking a flight, shopping, etc. The set of rules for extracting action parameters may specify extracting action parameters from a browsing intention and / or page content data of a task, etc. As an example, for an action type “searching” , corresponding action parameters may include a search query; for an action type “booking a hotel” , corresponding action parameters may include a destination, a check-in date, a check-out date, etc.; for an action type “booking a flight” , corresponding action parameters may include a departure city, a destination city, a departure date, etc.; for an action type “shopping” , corresponding action parameters may include a product name, a product model, etc.
[0055] Based on the prompt, the language model may extract the at least one action type and the at least one corresponding action parameter from the browsing intention and / or the page content data of the task, by utilizing its various capabilities such as a semantic understanding capability, a logical inference capability, etc. In this way, action types and action parameters can be accurately determined.
[0056] In an implementation, an action suggestion for a task may include a link to a webpage for performing a suggested action. The webpage may correspond to an action type and includes at least one action parameter corresponding to the action type. For example, a type of the webpage may be determined based on the action type. As an example, the webpage may be selected from predefined webpages with different types. Then an action parameter may be embedded into the webpage as a value of a corresponding variable of the webpage. For instance, if an action type is “shopping” and a corresponding action parameter includes a product name “keyboard” , a shopping webpage may be determined, and the product name “keyboard” may be embedded into the shopping webpage, as a search query. As such, with an action suggestion, a user may be directly navigated to a webpage for performing a suggested action, without complex interactive operations. Herein, the webpage for performing the action may be a webpage loaded in the browser or a webpage loaded in any other application other than the browser.
[0057] It should be understood that, the steps 242, 244, 246 and 260 may performed without utilizing a language model, may be performed through the same language model, or may be performed through different language models. The embodiments of the present disclosure are not limited in this regard. In an implementation, the same language model may be utilized for the steps 246 and 260, and the set of tasks 204 and the set of action suggestions 205 may be determined concurrently.
[0058] It should be appreciated that the process 200 is only exemplary, and according to actual application requirements and designs, various changes may be made to the process 200.
[0059] FIG. 3A illustrates an exemplary process 300A for generating visual representations corresponding to a set of tasks according to an embodiment.
[0060] As shown in FIG. 3A, the process 300A may include visual representation generation at 310. The visual representation generation at 310 may include generating visual representations 302 corresponding to a set of tasks 301. The set of tasks 301 may correspond to the set of tasks 103 shown in FIG. 1 or the set of tasks 204 shown in FIG. 2. Herein, a visual representation may refer to at least one user interface elements that can be presented on a user interface of a browser. Thus, the visual representations 302 may be loaded on a user interface of a browser for presenting to a user.
[0061] In an implementation, a visual representation corresponding to a task may include at least one user interface element for visually presenting the task on the user interface. As described above, in an implementation, a task may have a set of characteristics, such as a task name, a task category, page quantity of pages included in the task, a page title of each page, a page link of each page, a page browsing time point, etc. In this case, a visual representation corresponding to a task may include user interface elements respectively corresponding to a task name, a task category, page quantity, a page title of each page, a page browsing time point. For ease of description, herein, these user interface elements may be respectively referred to as a task-name element, a task-category element, page-quantity element, a page-title element and a page-browsing-time-point element. The page-title element may be loaded as a hyperlink to a corresponding page. The page-browsing-time-point element may be directly loaded as the page browsing time point, or may be loaded as indirectly indicating the page browsing time point. For example, the page-browsing-time-point element may be loaded as a length of time between current time point when the visual representation corresponding to the task is loaded and the page browsing time point, e.g., one day ago, 20 minutes ago, etc. Exemplary user interfaces will be described below with reference to FIG. 8.
[0062] Such visual representations enable users to intuitively view individual tasks via the user interface and effortlessly switch between different activities corresponding to different tasks.
[0063] FIG. 3B illustrates an exemplary process 300B for generating visual representations corresponding to a set of action suggestions according to an embodiment.
[0064] As shown in FIG. 3B, the process 300B may include visual representation generation at 350. The visual representation generation at 350 may include generating visual representations 352 corresponding to a set of action suggestions 351. The set of action suggestions 351 may correspond to the set of action suggestions 104 shown in FIG. 1 or the set of action suggestions 205 shown in FIG. 2. The visual representations 352 may be loaded on a user interface of a browser, thereby providing intuitive presentation of the action suggestions to users and efficiently assisting users in taking a corresponding action with simply interactive operations.
[0065] In an implementation, a visual representation corresponding to an action suggestion may include at least one user interface element for visually presenting the action suggestion on the user interface. For example, as described above, an action suggestion may include a link to a webpage for performing a suggested action. In this case, a visual representation corresponding to an action suggestion may include two user interface elements, which may be hereinafter referred to as an action-suggestion-indication element and action-suggestion element. The action-suggestion-indication element may indicate that an action suggestion is provided below this element, and the action-suggestion element may be presented as a webpage title of a webpage for performing a suggested action, and the webpage title may be loaded as a hyperlink to the webpage. Exemplary user interfaces will be described below with reference to FIG. 8.
[0066] It should be appreciated that the processes 300A and 300B are only exemplary, and according to actual application requirements and designs, various changes may be made to the process 300A or process 300B. For example, the visual representations 302 and the visual representations 352 may be generated together.
[0067] FIG. 4 illustrates an exemplary process 400 for generating and loading visual representations according to an embodiment. Through the process 400, visual representations corresponding to a set of tasks and / or a set of action suggestions may be generated and loaded on a user interface of a browser for presenting to a user.
[0068] As shown in FIG. 4, a request 401 for viewing a set of tasks may be received. The request 401 may be initiated by a user in any manners. For example, the request 401 may be initiated by a user through clicking on a user interface element for viewing the set of tasks. An exemplary user interface element for the request will be described below with reference to FIG. 8.
[0069] The process 400 may include visual representation generation at 420. The visual representation generation at 420 may include generating visual representations 402 corresponding to a set of tasks. For example, the visual representations 402 may be generated in a similar manner to the step 310 shown in FIG. 3A.
[0070] The process 400 may further include visual representation loading at 440. The visual representation loading at 440 may include loading the visual representations 402 on the user interface for presenting to a user. In this way, the user can intuitively view individual tasks via the user interface and effortlessly recall previous activities corresponding to the tasks and quickly switch between different activities.
[0071] Optionally, if a set of action suggestions has been determined for a task of the set of tasks as described above, the visual representation generation at 420 may further include generating visual representations corresponding to the set of action suggestions. For example, the visual representations corresponding to the set of action suggestions may be generated in a similar manner to the step 350 shown in FIG. 3B. Accordingly, the visual representation loading at 440 may further include loading the visual representations corresponding to the set of action suggestions on the user interface for presenting to the user. Preferably, the visual representations corresponding to the set of action suggestions may be loaded together with the visual representation of the corresponding task. The intuitive presentation of the set of action suggestions on the user interface can efficiently assist users in taking a corresponding action with simply interactive operations.
[0072] It should be appreciated that the process 400 is only exemplary, and according to actual application requirements and designs, various changes may be made to the process 400.
[0073] FIG. 5 illustrates an exemplary process 500 for task searching and visual representation loading according to an embodiment. In the process 500, in response to a task query, a task matching the task query may be determined and a corresponding visual representation may be generated and loaded on a user interface of a browser.
[0074] As shown in FIG. 5, a task query 501 may be received. For example, the task query 501 may be input by a user via a search area (e.g., a search box or bar) presented on the user interface. The task query 501 may have any modality, such as text, audio, image, etc. Task query 501 may indicate that the user wants to find a desired task from a set of tasks 502.
[0075] The process 500 may include task search at 520. The task search at 520 may include, based on the task query 501, performing a search operation on a set of tasks 502 to obtain at least one matching task 503 which matches the task query 501. The set of tasks 502 may correspond to the set of tasks 103 shown in FIG. 1 or the set of tasks 204 shown in FIG. 2.
[0076] In an implementation, the task search 520 may include selecting the at least one matching task 503 based on relevance between the task query 501 and each task of the set of tasks 502. The relevance may be determined by using any suitable techniques such as a text embedding technique. For example, the task query 501 may be encoded as a query embedding, and a characteristic (such as a task name, a page title, etc. ) of each task of the set of tasks 502 may be encoded as a task embedding. Then a relevance score between the query embedding and the task embedding may be calculated, and the relevance score may indicate the relevance between the task query 501 and the task. At least one task with the highest relevance scores may be selected as the at least one matching task 503 which matches the task query 501.
[0077] The process 500 may further include visual representation generation at 540. visual representation generation at 540 may include generating at least one visual representation 504 corresponding to the at least one matching task 503. For example, the at least one visual representation 504 corresponding to the at least one matching task 503 may include user interface elements for representing the at least one matching task 503. The visual representation generation at 540 may be implemented in a similar manner to the step 310 shown in FIG. 3A.
[0078] The process 500 may further include visual representation loading at 560. The visual representation loading at 560 may include loading the at least one visual representation 504 on the user interface for presenting to a user.
[0079] In an implementation, the process 500 may be performed after the process 400. For example, when a user is viewing, on the user interface, a set of tasks which is loaded and presented according to the process 400, the user may initiate to search for a target task by inputting a task query on the user interface, and this task query may trigger the process 500 to return a search result, such as, loading the at least one visual representation 504 on the user interface. Through the process 500, a user can swiftly locate a target task and accordingly locate pages included in this task, thereby significantly saving time in searching for the pages.
[0080] It should be appreciated that the process 500 is only exemplary, and according to actual application requirements and designs, various changes may be made to the process 500. In an implementation, if a set of action suggestions has been generated for the matching task 503, the visual representation generation at 540 may further include generating visual representations corresponding to the set of action suggestions, and the visual representation loading at 560 may further include loading the visual representations corresponding to the set of action suggestions on the user interface. In such an implementation, the visual representations corresponding to the set of action suggestions may be generated in a similar manner to the step 350 shown in FIG. 3B.
[0081] FIG. 6 illustrates an exemplary process 600 for page resumption according to an embodiment. Through the process 600, pages included in a task may be loaded together on a user interface of a browser in response to an interactive operation on a visual representation corresponding to the task.
[0082] As shown in FIG. 6, the process 600 may include interactive operation detection at 620. The interactive operation detection at 620 may include detecting an interactive operation 602 on a visual representation 601 corresponding to a task. The interactive operation 602 may be initiated by a user on the user interface. For example, the user may click on the visual representation 601, which may be taken as the interactive operation 602.
[0083] The process 600 may further include page loading at 640. The page loading at 640 may include, in response to the interactive operation 602, loading pages included in the task on the user interface for presenting to the user. For example, the pages may be concurrently loaded on the user interface.
[0084] The visual representation 601 may correspond to one of the visual representations 402 shown in FIG. 4 or the visual representation 504 shown in FIG. 5. Accordingly, the process 600 may be performed after the process 400 or the process 500. As an example, when visual representations corresponding to a set of tasks are loaded and presented on a user interface, in response to an interactive operation on a visual representation corresponding to a task, all the pages included in the task can be reopened in the browser. As another example, if a user is searching for a target task by inputting a task query on the user interface, and the target task has been found, all the pages included in the target task would be reopened in the browser in response to an interactive action on a visual representation corresponding to the target task on the user interface.
[0085] Through the process 600, all pages included in a task can be reopened together just in response to an interactive operation on a visual representation of the task, thus, the efficiency of page resumption can be significantly improved and user interactive operations can be greatly reduced, thereby providing more seamless and less stressful user browsing experience.
[0086] It should be appreciated that the process 600 is only exemplary, and according to actual application requirements and designs, various changes may be made to the process 600.
[0087] FIG. 7 illustrates an exemplary process 700 for page quick launch according to an embodiment. In the process 700, with quick launch options corresponding to tasks, pages included in a task may be loaded quickly on a user interface of a browser.
[0088] As shown in FIG. 7, the process 700 may include quick launch option loading at 720. The quick launch option loading at 720 may include loading at least one quick launch option on the user interface. Each quick launch option of the at least one quick launch option may correspond to a task of a set of tasks, wherein the set of tasks may be the set of tasks 103 shown in FIG. 1 or the set of tasks 204 shown in FIG. 2.
[0089] The at least one quick launch option may be loaded at any suitable position on the user interface and / or may be loaded in response to any trigger. As an example, when an interactive operation on an address bar is detected, the at least one quick launch option may be loaded in a dropdown menu of the address bar. As another example, the at least one quick launch option may be loaded in a sidebar, and the sidebar may be loaded at one side of the user interface.
[0090] The process 700 may further include interactive operation detection at 740. The interactive operation detection at 740 may include detecting an interactive operation 702 on a quick launch option 701 of the at least one quick launch option. For example, the user may click on the quick launch option 701, which may be taken as the interactive operation 702.
[0091] The process 700 may further include page loading at 760. The page loading at 760 may include, in response to the interactive operation 702, loading pages included in a task corresponding to the quick launch option 701 on the user interface for presenting to the user.
[0092] Through the process 700, a user may reopen pages included in a corresponding task in a quick and convenient manner, thereby further enhancing user browsing experience and satisfaction.
[0093] It should be appreciated that the process 700 is only exemplary, and according to actual application requirements and designs, various changes may be made to the process 700.
[0094] FIG. 8 illustrates exemplary user interfaces 800A and 800B according to an embodiment. The user interface 800A depicts an example of a user interface element for a request for viewing tasks (e.g., corresponding to the request 401 shown in FIG. 4) , and the user interface 800B depicts an example of visual representations corresponding to tasks (e.g., corresponding to the visual representations 302 shown in FIG. 3A) and visual representations corresponding to action suggestions (e.g., corresponding to the visual representations 352 shown in FIG. 3B) .
[0095] As shown in FIG. 8, the user interface 800A may be, e.g., a user interface after a browser has been launched. A tab bar 801 is loaded at the top of the user interface 800A, and the tab bar 801 currently contains a new tab 802. An address bar 803 is also loaded below the tab bar 801, and the address bar 803 is used for receiving a webpage address input by a user or presenting a webpage address of a currently loaded webpage. For example, the webpage address may be a Universal Resource Locator (URL) .
[0096] On the user interface 800A, a task-center element 804 is loaded on one side (e.g., the right-hand side) of the address bar 803. Via the task-center element 804, a request for viewing a set of tasks (such as the set of tasks 103 shown in FIG. 1 or the set of tasks 204 shown in FIG. 2) may be initiated. For example, a user may click on the task-center element 804 so as to initiate the request for viewing the set of tasks.
[0097] The user interface 800B may be, e.g., a user interface after the user has clicked on the task-center element 804. On the user interface 800B, the tab bar 801 contains a tab 805 titled “Task center” , which may indicate that visual representations corresponding to the set of tasks and visual representations corresponding to action suggestions are loaded in a main window below the tab bar 801. Additionally, a search bar 806 is loaded below the tab bar 801, and the search bar 806 may be used for receiving a task query, which will be described in detail with reference to FIG. 9.
[0098] As shown in the user interface 800B, the visual representations corresponding to the set of tasks and the visual representations corresponding to action suggestions may be presented as cards. As an example, cards 810, 820, 830, etc., are included in the user interface 800B. Cards 810, 820, 830 respectively correspond to three exemplary tasks and corresponding action suggestions. For example, the three exemplary tasks and the corresponding action suggestions may be generated according to the process 100 shown in FIG. 1 or the process 200 shown in FIG. 2, and the cards 810, 820 and 830 may be generated according to the process 300A shown in FIG. 3A and the process 300B shown in FIG. 3B.
[0099] The card 810 may contain user interface elements for a corresponding first task, including: such as, a task-category element 811 (e.g., “Shopping” ) ; a task-name element 812 (e.g., “Buy a router” ) ; a page-quantity and page-browsing-time-point element 813 (e.g., “4 pages, 1 day ago” ) ; page-title elements 814 (e.g., “Router -Search” , “Wired Router” , “Wireless Router” , “Router Rankings” ) ; and so on. The page-title elements 814 may be loaded as hyperlinks to respective pages.
[0100] The card 810 may also contain visual representations corresponding to an action suggestion for the corresponding first task. Such visual representations may include an action-suggestion-indication element 815 (e.g., “Explore Further” ) and an action-suggestion element 816 (e.g., “Shopping for router” ) . The action-suggestion-indication element 815 may indicate that an action suggestion is provided below, and the action-suggestion element 816 may be loaded as a hyperlink to a webpage for shopping for a router.
[0101] Similarly, the card 820 may contain user interface elements for a corresponding second task, including, such as, a task-category element 821, a task-name element 822, a page-quantity and page-browsing-time-point element 823, page-title elements 824, etc. Additionally, the card 820 may contain a three-dot element 825 which may indicate that there are more page titles available. The card 820 may also contain an action-suggestion-indication element 826 and an action-suggestion element 827.
[0102] Similarly, the card 830 may contain user interface elements for a corresponding third task, including, such as, a task-category element 831, a task-name element 832, a page-quantity and page-browsing-time-point element 833, page-title elements 834, a three-dot element 835, an action-suggestion-indication element 835 and two action-suggestion elements 837 and 838.
[0103] It should be appreciated that the user interfaces 800A and 800B are only exemplary, and according to actual application requirements and designs, the user interfaces may have any other layouts and may include more or fewer elements. For example, the task-center element 804 may have any other form or may be placed at any other position on the user interface. For another example, a card may contain more or less elements and may present the elements in any other forms.
[0104] FIG. 9 illustrates exemplary user interfaces 900A and 900B according to an embodiment. The user interfaces 900A and 900B depict examples of inputting a task query and loading a corresponding search result.
[0105] The user interface 900A may correspond to the user interface 800B shown in FIG. 8. Three cards 910, 920 and 930 corresponding to three tasks may be loaded in a main window of the user interface 900A. The cards 910, 920 and 930 may correspond to the cards 810, 820 and 830 in FIG. 8 respectively. Additionally, a search bar 901 is loaded above the cards. A user is currently inputting a task query in the search bar 901.
[0106] The user interfaces 900B may be a user interface after the user has input the task query 902 “Router” in the search bar 901 and a matching task has been found. At this time, only the card 910 is loaded in a main window of the user interface 900B, wherein a task corresponding to the card 910 is determined as matching the task query 902. For example, the card 910 may be loaded according to the process 500 shown in FIG. 5.
[0107] It should be appreciated that the user interfaces 900A and 900B are only exemplary, and according to actual application requirements and designs, the user interfaces may have any other layouts and may include more or fewer elements.
[0108] FIG. 10 illustrates an exemplary user interface 1000 according to an embodiment. The user interface 1000 depicts an example of page resumption.
[0109] As shown in FIG. 10, the user interface 1000 may be a user interface after a user has clicked on the card 810 shown in FIG. 8 or the card 910 shown in FIG. 9. On the user interface 1000, a tab bar contains four tabs 1010, 1020, 1030 and 1040 corresponding to four pages included in the task corresponding to the card 810 or 910. This means that the four pages have been loaded on the user interface 1000. The loading of the four pages may be implemented, for example, according to the process 600 shown in FIG. 6.
[0110] The tab 1010 is currently active, and accordingly, a page corresponding to the tab 1010 is currently loaded in a main window of the user interface 1000. The page may be a search result webpage. An address bar 1011 may contain an address of this page. A search bar 1012 may contain a search query “Router” and some search results for the search query “Router” are loaded in an area 1013.
[0111] It should be appreciated that the user interface 1000 is only exemplary, and according to actual application requirements and designs, the user interface may have any other layouts and may include more or fewer elements.
[0112] FIG. 11 illustrates an exemplary user interface 1100 according to an embodiment. The user interface 1100 depicts an example of a webpage for performing an action.
[0113] As shown in FIG. 11, the user interface 1100 may be a user interface after a user has clicked on the action-suggestion element 816 shown in FIG. 8. As described in combination with FIG. 8, the action-suggestion element 816 may be a hyperlink to a webpage for shopping for a router. Thus, after the user has clicked on the action-suggestion element 816, the webpage for shopping for a router may be loaded, such as, a webpage corresponding to a tab 1110. An address bar 1111 may contain an address of the webpage. The webpage may contain a search bar 1112, in which a query “Router” has been loaded. Some search results matching the query “Router” are loaded in an area 1113.
[0114] It should be understood that, the webpage may be a webpage loaded via the browser, or may be a webpage loaded via another application different from the browser, and the embodiments of the present disclosure are not limited in this regard. Such a webpage may be generated, for example, in a similar manner as described for the step 260 shown in FIG. 2.
[0115] It should be appreciated that the user interface 1100 is only exemplary, and according to actual application requirements and designs, the user interface may have any other layouts and may include more or fewer elements.
[0116] FIG. 12 illustrates exemplary user interfaces 1200A and 1200B according to an embodiment. The user interfaces 1200A or 1200B depict examples of at least one quick launch option.
[0117] The user interface 1200A may be a user interface after a browser is launched. When the user clicks on an address bar, a dropdown menu 1210 of the address bar may be loaded. The dropdown menu 1210 may contain at least one quick launch option, such as, quick launch options 1211 and 1212. As an example, the quick launch option 1211 may correspond to the same task as the card 810 shown in FIG. 8, and the quick launch option 1212 may correspond to the same task as the card 820 shown in FIG. 8.
[0118] The user interface 1200B may also be a user interface after a browser is launched. Different from the user interface 1200A, on the user interface 1200B, at least one quick launch option, such as quick launch options 1252 and 1253, is loaded in a sidebar 1250. As an example, the quick launch option 1252 may correspond to the same task as the card 810 shown in FIG. 8, and the quick launch option 1253 may correspond to the same task as the card 820 shown in FIG. 8. The sidebar 1250 may also contain a search bar 1251. Although the sidebar 1250 is shown as being placed on the right-hand side of the user interface 1200B, the sidebar 1250 may also be placed at any position of the user interface 1200B.
[0119] The quick launch options 1211 and 1212 or the quick launch options 1252 and 1253 may be loaded, for example, in a similar manner to the step 720 shown in FIG. 7.
[0120] As described above, a quick launch option may provide a quick manner of reopening pages included in a task corresponding to the quick launch option. Thus, when the user clicks on a quick launch option, corresponding pages will be loaded on the user interface. For example, after the user has clicked on the quick launch option 1211 or the quick launch option 1252, the user interface shown in FIG. 11 may be loaded for presenting to the user.
[0121] It should be appreciated that the user interfaces 1200A and 1200B are only exemplary, and according to actual application requirements and designs, the user interfaces may have any other layouts and may include more or fewer elements. For example, the quick launch options may be placed at any positions on the user interface, or may have any forms, or any quantity of quick launch options may be loaded.
[0122] FIG. 13 illustrates a flowchart of an exemplary method 1300 for browsing history management according to an embodiment.
[0123] At 1310, raw data of browsing history may be retrieved.
[0124] At 1320, reference data may be obtained. The reference data may include at least one portion of the raw data, and the at least one portion of the raw data may include a record of webpages being browsed within a predetermined period of time.
[0125] At 1330, content data of each webpage may be obtained through a language model.
[0126] At 1340, browsing intention for each webpage may be extracted, through the language model, from content data of the webpage.
[0127] At 1350, a set of tasks may be generated through the language model. Each task may include at least two pages, and the at least two pages may be a subset of the webpages and associated with the same browsing intention.
[0128] In an implementation, the record may include webpage information and interaction information of the webpages.
[0129] In an implementation, each task may be represented by at least one of: task name, task category, page quantity, page titles, page links, page browsing time point, and page browsing duration.
[0130] In an implementation, the method 1300 may further comprise: determining, through the language model, a set of action suggestions for a task, based on at least one of: browsing intention of the task, and page content data of the task.
[0131] The determining, through the language model, a set of action suggestions for a task may comprise: extracting, through the language model, at least one action type, from the browsing intention and / or the page content data; extracting, through the language model, at least one corresponding action parameter, from the browsing intention and / or the page content data; and generating the set of action suggestions based on the at least one action type and the at least one corresponding action parameter.
[0132] Each action suggestion may include a link to a webpage, and the webpage may correspond to an action type and include a corresponding action parameter.
[0133] The method 1300 may further comprise: receiving a request for viewing the set of tasks; and in response to the request, loading visual representations corresponding to the set of tasks and visual representations corresponding to the set of action suggestions, on a user interface of a browser.
[0134] In an implementation, the method 1300 may further comprise: receiving a request for viewing the set of tasks; and in response to the request, loading visual representations corresponding to the set of tasks, on a user interface of a browser.
[0135] The method 1300 may further comprise: receiving a task query; in response to the task query, determining at least one task matching the task query; and loading at least one visual representation corresponding to the at least one task, on the user interface.
[0136] The method 1300 may further comprise: detecting an interactive operation on a visual representation corresponding to a task; and in response to the interactive operation, loading pages included in the task on the user interface.
[0137] In an implementation, the method 1300 may further comprise: loading at least one quick launch option on a user interface of a browser, each quick launch option corresponding to a task; detecting an interactive operation on a quick launch option; and in response to the interactive operation, loading pages included in a task corresponding to the quick launch option, on the user interface.
[0138] It should be appreciated that the method 1300 may further comprise any steps / processes for browsing history management according to the embodiments of the present disclosure as mentioned above.
[0139] FIG. 14 illustrates an exemplary apparatus 1400 for browsing history management according to an embodiment.
[0140] The apparatus 1400 may comprise: a raw data retrieving module 1410, for retrieving raw data of browsing history; a reference data obtaining module 1420, for obtaining reference data, the reference data including at least one portion of the raw data, and the at least one portion of the raw data including a record of webpages being browsed within a predetermined period of time; a content data obtaining module 1430, for obtaining, through a language model, content data of each webpage; a browsing intention extracting module 1440, for extracting, through the language model, browsing intention for each webpage, from the content data; and a task generating module 1450, for generating, through the language model, a set of tasks, wherein each task includes at least two pages, the at least two pages being a subset of the webpages, and associated with the same browsing intention. Moreover, the apparatus 1400 may also comprise any other modules configured for performing any steps and operations of the methods for browsing history management according to the embodiments of the present disclosure as mentioned above.
[0141] FIG. 15 illustrates an exemplary apparatus 1500 for browsing history management according to an embodiment.
[0142] The apparatus 1500 may comprise at least one processor 1510 and a memory 1520 storing computer-executable instructions. When the computer-executable instructions are executed, the at least one processor 1510 may: retrieve raw data of browsing history; obtain reference data, the reference data including at least one portion of the raw data, the at least one portion of the raw data including a record of webpages being browsed within a predetermined period of time; obtain, through a language model, content data of each webpage; extract, through the language model, browsing intention for each webpage, from the content data; and generate, through the language model, a set of tasks, wherein each task includes at least two pages, the at least two pages being a subset of the webpages and associated with the same browsing intention. The at least one processor 1510 may be further configured for performing any operations of the methods for browsing history management according to the embodiments of the present disclosure as mentioned above.
[0143] The embodiments of the present disclosure propose a computer program product for browsing history management. The computer program product may comprise a computer program that is executed by at least one processor for: retrieving raw data of browsing history; obtaining reference data, the reference data including at least one portion of the raw data, the at least one portion of the raw data including a record of webpages being browsed within a predetermined period of time; obtaining, through a language model, content data of each webpage; extracting, through the language model, browsing intention for each webpage, from the content data; and generating, through the language model, a set of tasks, wherein each task includes at least two pages, the at least two pages being a subset of the webpages and associated with the same browsing intention. The computer program may be further executed by the at least one processor for performing any operations of the methods for browsing history management according to the embodiments of the present disclosure as mentioned above.
[0144] The embodiments of the present disclosure may be embodied in a non-transitory computer-readable medium. The non-transitory computer-readable medium may comprise instructions that, when executed, cause one or more processors to perform any operations of the methods for browsing history management according to the embodiments of the present disclosure as mentioned above.
[0145] It should be appreciated that all the operations in the methods described above are merely exemplary, and the present disclosure is not limited to any operations in the methods or sequence orders of these operations, and should cover all other equivalents under the same or similar concepts.
[0146] It should also be appreciated that all the modules in the apparatuses described above may be implemented in various approaches. These modules may be implemented as hardware, software, or a combination thereof. Moreover, any of these modules may be further functionally divided into sub-modules or combined together.
[0147] Processors have been described in connection with various apparatuses and methods. These processors may be implemented using electronic hardware, computer software, or any combination thereof. Whether such processors are implemented as hardware or software will depend upon the particular application and overall design constraints imposed on the system. By way of example, a processor, any portion of a processor, or any combination of processors presented in the present disclosure may be implemented with a microprocessor, microcontroller, digital signal processor (DSP) , a field-programmable gate array (FPGA) , a programmable logic device (PLD) , a state machine, gated logic, discrete hardware circuits, and other suitable processing components configured to perform the various functions described throughout the present disclosure. The functionality of a processor, any portion of a processor, or any combination of processors presented in the present disclosure may be implemented with software being executed by a microprocessor, microcontroller, DSP, or other suitable platform.
[0148] Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, threads of execution, procedures, functions, etc. The software may reside on a computer-readable medium. A computer-readable medium may include, by way of example, memory such as a magnetic storage device (e.g., hard disk, floppy disk, magnetic strip) , an optical disk, a smart card, a flash memory device, random access memory (RAM) , read only memory (ROM) , programmable ROM (PROM) , erasable PROM (EPROM) , electrically erasable PROM (EEPROM) , a register, a removable disk, or a solid state drive. Although memory is shown separate from the processors in the various aspects presented throughout the present disclosure, the memory may be internal to the processors, e.g., cache or register.
[0149] Moreover, the articles “a” and “an” as used in this specification and the appended claims should generally be construed to mean “one” or “one or more” unless specified otherwise or clear from the context to be directed to a singular form.
[0150] The previous description is provided to enable any person skilled in the art to practice the various aspects described herein. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects. Thus, the claims are not intended to be limited to the aspects shown herein. All structural and functional equivalents to the elements of the various aspects described throughout the present disclosure that are known or later come to be known to those of ordinary skilled in the art are intended to be encompassed by the claims.
Claims
1.A method for browsing history management, comprising:retrieving raw data of browsing history;obtaining reference data, the reference data including at least one portion of the raw data, the at least one portion of the raw data including a record of webpages being browsed within a predetermined period of time;obtaining, through a language model, content data of each webpage;extracting, through the language model, browsing intention for each webpage, from the content data; andgenerating, through the language model, a set of tasks, wherein each task includes at least two pages, the at least two pages being:a subset of the webpages, andassociated with the same browsing intention.2.The method of claim 1, whereinthe record includes webpage information and interaction information of the webpages.3.The method of claim 1, whereineach task is represented by at least one of: task name, task category, page quantity, page titles, page links, page browsing time point, and page browsing duration.4.The method of claim 1, further comprising:determining, through the language model, a set of action suggestions for a task, based on at least one of:browsing intention of the task, andpage content data of the task.5.The method of claim 4, wherein the determining, through the language model, a set of action suggestions for a task comprises:extracting, through the language model, at least one action type, from the browsing intention and / or the page content data;extracting, through the language model, at least one corresponding action parameter, from the browsing intention and / or the page content data; andgenerating the set of action suggestions based on the at least one action type and the at least one corresponding action parameter.6.The method of claim 5, whereineach action suggestion includes a link to a webpage, and the webpage corresponds to an action type and includes a corresponding action parameter.7.The method of claim 4, further comprising:receiving a request for viewing the set of tasks; andin response to the request, loading visual representations corresponding to the set of tasks and visual representations corresponding to the set of action suggestions, on a user interface of a browser.8.The method of claim 1, further comprising:receiving a request for viewing the set of tasks; andin response to the request, loading visual representations corresponding to the set of tasks, on a user interface of a browser.9.The method of claim 8, further comprising:receiving a task query;in response to the task query, determining at least one task matching the task query; andloading at least one visual representation corresponding to the at least one task, on the user interface.10.The method of any of claims 7 to 9, further comprising:detecting an interactive operation on a visual representation corresponding to a task; andin response to the interactive operation, loading pages included in the task on the user interface.11.The method of claim 1, further comprising:loading at least one quick launch option on a user interface of a browser, each quick launch option corresponding to a task;detecting an interactive operation on a quick launch option; andin response to the interactive operation, loading pages included in a task corresponding to the quick launch option, on the user interface.12.An apparatus for browsing history management, comprising:at least one processor; anda memory storing computer-executable instructions that, when executed, cause the at least one processor to:retrieve raw data of browsing history;obtain reference data, the reference data including at least one portion of the raw data, the at least one portion of the raw data including a record of webpages being browsed within a predetermined period of time;obtain, through a language model, content data of each webpage;extract, through the language model, browsing intention for each webpage, from the content data; andgenerate, through the language model, a set of tasks, wherein each task includes at least two pages, the at least two pages being:a subset of the webpages, andassociated with the same browsing intention.13.The apparatus of claim 12, whereineach task is represented by at least one of: task name, task category, page quantity, page titles, page links, page browsing time point, and page browsing duration.14.The apparatus of claim 12, wherein the computer-executable instructions, when executed, further cause the at least one processor to:determine, through the language model, a set of action suggestions for a task, based on at least one of: browsing intention of the task, and page content data of the task.15.The apparatus of claim 14, wherein the computer-executable instructions, when executed, further cause the at least one processor to:receive a request for viewing the set of tasks; andin response to the request, load visual representations corresponding to the set of tasks and visual representations corresponding to the set of action suggestions, on a user interface of a browser.16.The apparatus of claim 12, wherein the computer-executable instructions, when executed, further cause the at least one processor to:receive a request for viewing the set of tasks; andin response to the request, load visual representations corresponding to the set of tasks, on a user interface of a browser.17.The apparatus of claim 16, wherein the computer-executable instructions, when executed, further cause the at least one processor to:receive a task query;in response to the task query, determine at least one task matching the task query; andload at least one visual representation corresponding to the at least one task, on the user interface.18.The apparatus of any of claims 15 to 17, wherein the computer-executable instructions, when executed, further cause the at least one processor to:detect an interactive operation on a visual representation corresponding to a task; andin response to the interactive operation, load pages included in the task on the user interface.19.The apparatus of claim 12, wherein the computer-executable instructions, when executed, further cause the at least one processor to:load at least one quick launch option on a user interface of a browser, each quick launch option corresponding to a task;detect an interactive operation on a quick launch option; andin response to the interactive operation, load pages included in a task corresponding to the quick launch option, on the user interface.20.A computer program product for browsing history management, the computer program product comprising a computer program that is executed by at least one processor for:retrieving raw data of browsing history;obtaining reference data, the reference data including at least one portion of the raw data, the at least one portion of the raw data including a record of webpages being browsed within a predetermined period of time;obtaining, through a language model, content data of each webpage;extracting, through the language model, browsing intention for each webpage, from the content data; andgenerating, through the language model, a set of tasks, wherein each task includes at least two pages, the at least two pages being:a subset of the webpages, andassociated with the same browsing intention.