Network link processing method and device, equipment and storage medium

Abstract

The invention provides a network link processing method and device, equipment and a storage medium, and relates to the technical field of computers, in particular to the technical field of intelligent search. According to the specific implementation scheme, the method comprises the steps of obtaining a statistical record of a target link set in at least one statistical period; the statistical record is used for recording network communication error conditions of a plurality of communication operation objects; and based on the statistical record, adjusting a recommendation sequence of the target links in the target link set in the corresponding to-be-recommended resource set.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and more particularly to the field of intelligent search technology. Background Technology

[0002] Web links (such as URLs, or Uniform Resource Locators) can locate and access web resources. Web resources include, but are not limited to, web pages, audio, and video. Content providers can publish web resources through web links, and users can access web resources through web links. For example, users can use a browser to access web pages, or use applications installed on their devices to browse online stores and watch online videos.

[0003] In real-world applications, various factors can cause network connectivity issues, preventing users from accessing network resources. Sending these problematic links to users can negatively impact their user experience. Summary of the Invention

[0004] This disclosure provides a method, apparatus, device, and storage medium for processing network links.

[0005] According to one aspect of this disclosure, a method for processing network links is provided, comprising:

[0006] Obtain statistical records of the target link set within at least one statistical period; the statistical records are used to record network communication error conditions of multiple communication operation objects;

[0007] Based on the statistical records, the recommendation ranking of the target links in the target link set is adjusted in the corresponding resource set to be recommended.

[0008] According to another aspect of this disclosure, a network link processing apparatus is provided, comprising:

[0009] The acquisition module is used to acquire statistical records of the target link set within at least one statistical period; the statistical records are used to record network communication error conditions of multiple communication operation objects;

[0010] The adjustment module is used to adjust the recommendation ranking of target links in the target link set within the corresponding resource set to be recommended, based on the statistical records.

[0011] According to another aspect of this disclosure, an electronic device is provided, comprising:

[0012] At least one processor; and

[0013] The memory is communicatively connected to the at least one processor; wherein,

[0014] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform any of the methods described in the present disclosure.

[0015] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are used to cause the computer to perform any of the methods according to embodiments of this disclosure.

[0016] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements any of the methods according to embodiments of this disclosure.

[0017] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0018] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0019] Figure 1 This is a schematic diagram illustrating an application scenario of the network link processing method according to the first embodiment of this disclosure;

[0020] Figure 2 This is a flowchart illustrating a method for processing network links according to a second embodiment of the present disclosure;

[0021] Figure 3 This is a flowchart illustrating a method for processing network links according to a third embodiment of this disclosure;

[0022] Figure 4 This is a flowchart illustrating a method for processing network links according to the fourth embodiment of this disclosure;

[0023] Figure 5 This is a schematic diagram of the structure of a network link processing apparatus according to the fifth embodiment of this disclosure;

[0024] Figure 6 This is a block diagram of an electronic device used to implement the network link processing method of the embodiments of this disclosure. Detailed Implementation

[0025] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0026] The terms “first,” “second,” etc., used in this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion, such as including a series of steps or units. A method, system, product, or apparatus is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products, or apparatuses.

[0027] It should be noted that, unless it is explicitly stated that there is a sequential order of execution between different operations, or that there is a sequential order of execution between different operations in terms of technical implementation, the execution order between multiple operations may not be significant, and multiple operations may be executed simultaneously.

[0028] Among related technologies, web crawling can be used to discover dead links. Dead links are links that were originally accessible but have since become invalid. These links prevent users from accessing the content related to those links on the web. Therefore, web crawling technology can, to a certain extent, identify problematic web links.

[0029] However, web crawling technology is often used to analyze a fixed set of web links, and once a link encounters a problem, it is considered a dead link and no longer pushed to users. This means that web crawling technology cannot be applied to a large number of newly added links. Moreover, the network environment is complex; a web link that is dead in network operator A's environment may be a normally accessible web link in network operator B's environment.

[0030] In view of this, embodiments of the present disclosure provide a method for processing network links. This method uses a statistical approach to effectively distinguish network communication error situations of different network operators, flexibly identify potential dead links, and adjust the recommended ranking of problematic network links based on statistical results, rather than simply not recommending them.

[0031] like Figure 1 The diagram shown illustrates an application scenario of the network link processing method provided in this embodiment. Figure 1 It includes server 11 and terminal device 12.

[0032] The terminal device 12 and the server 11 are connected via a wireless or wired network. The terminal device 12 includes, but is not limited to, electronic devices such as desktop computers, mobile phones, portable computers, tablets, media players, smart wearable devices, and smart TVs. The server 11 can be a single server, a server cluster consisting of several servers, or a cloud computing center. The server 11 can be an independent physical server, a server cluster consisting of multiple physical servers, or a distributed system. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0033] The network link processing method in this embodiment of the disclosure can be provided by... Figure 1 The server shown performs the following: The server adjusts the recommended ranking of problematic network links according to the network link processing method provided in this embodiment of the disclosure, thereby reducing the likelihood of recommending problematic network links to users and improving user experience.

[0034] like Figure 2 The diagram shown is a flowchart of the method, including the following:

[0035] S201, Obtain statistical records of the target link set within at least one statistical period; the statistical records are used to record network communication error conditions of multiple communication operation objects.

[0036] The target link set refers to the set obtained by clustering multiple network links according to certain rules. The target link set may include at least one network link.

[0037] The statistical period can be set based on actual conditions, such as one day, three days, or seven days. Of course, the statistical period can also be shorter than one day; it can be flexibly set according to actual needs during implementation.

[0038] The communication operation target mainly refers to network operators. Network communication errors can include many types, such as connection errors, data transmission errors, protocol errors, network configuration errors, etc., and each error type includes multiple subtypes. The embodiments of this disclosure mainly focus on optimizing the recommendation ranking for potential dead links, so the network communication errors analyzed mainly include at least one of network layer errors, HTTP errors, rendering errors, etc.

[0039] S202, Based on statistical records, adjust the recommendation ranking of target links within the target link set in the corresponding resource set to be recommended.

[0040] During implementation, the recommendation ranking of problematic or abnormal target links in the set of resources to be recommended is adjusted backward, and then the recommendation ranking optimized based on statistical records is recommended to the target object.

[0041] In this embodiment, statistical records of the target link set over at least one statistical period are recorded, enabling targeted optimization of the recommendation ranking of target links within the target link set based on actual network communication error data. This allows for priority display of normal links during subsequent recommendation processes, while potentially abnormal network links (such as potentially dead links) are ranked lower, minimizing the exposure rate of potentially dead links and reducing the likelihood of recommending problematic network links. This further enhances user experience and improves the rationality of resource allocation and the effectiveness of recommendations. Furthermore, statistical analysis of a set of target links reduces the consumption of computational resources due to the large amount of network link resources and mitigates the problem of unreliable statistical results caused by insufficient statistical samples of individual network link resources, thereby improving the rationality of recommendation ranking adjustments. Simultaneously, performing statistical analysis and recommendation ranking adjustments on a set-by-set basis improves processing efficiency.

[0042] In some embodiments, in order to accurately count the potential dead links of different communication operators, this disclosure proposes to generate statistical records based on user posterior signals and multi-dimensional statistical analysis, starting from the terminal device.

[0043] In practice, each time a terminal device initiates a network connection access request (e.g., a webpage loading request initiated through a browser or mobile terminal app), the terminal device can track and analyze the processing status of the access request to generate log information and report it to the server.

[0044] For servers, it can be based on Figure 3 The method shown is used to generate statistical records of the target link set for each statistical period:

[0045] S301, Obtain log information generated by multiple terminal devices within the statistical period; the log information includes multiple access records; each access record includes: a link to be processed, a classification tag corresponding to the link to be processed, and a communication operation object identifier; the classification tag is used to indicate the network communication status corresponding to the link to be processed.

[0046] During implementation, the target device initiates an access request for the link to be processed through a terminal device. The terminal device can monitor the following key nodes throughout the process based on this access request to statistically analyze network communication errors:

[0047] 1) Network layer errors, including error types such as those in the Chromium NetErrorList (web browser network error list). Examples include: DNS (Domain Name System) resolution failure, TCP (Transmission Control Protocol) connection reset, TCP connection refused, etc.

[0048] Among them, DNS resolution failure (-105): Turbonet will identify DNS resolution failure errors if the requested domain name cannot be correctly resolved to an IP address during page loading.

[0049] TCP Connection Reset (-101): This is used to detect TCP connection reset errors if the TCP connection is reset by the other party during page loading. Turbonet will recognize the TCP connection reset error.

[0050] TCP connection refused (-102): This is used to identify TCP connection refused errors during page loading.

[0051] Other error types are not listed in this embodiment.

[0052] 2) HTTP errors, such as 40x, 50x and response timeout (e.g., no response for more than 6 seconds).

[0053] Error 40x is usually related to client-side requests. Check the request format, path, permissions, etc.

[0054] Error 50x: Usually related to the server side. Check server configuration, load, logs, etc.

[0055] 3) Rendering error: As the name suggests, the terminal device failed to render the content of the network link. This could be due to errors in the page's code, causing the content on the page to fail to display.

[0056] The above types of errors are tracked and analyzed by the terminal device, generating corresponding access records in the log information. During implementation, the terminal device can use the Turbonet network module to proxy requests and track and record access to the pending links that need to be loaded.

[0057] In order to optimize the recommended ranking of network links and reduce the storage and network transmission resources consumed by log information, the access records of log information in this embodiment of the disclosure, as described above, mainly include links to be processed, classification tags corresponding to the links to be processed, and communication operation object identifiers.

[0058] The communication operator object identifier can be obtained by the terminal device through reverse lookup of the IP (Internet Protocol) address of the link to be processed, or directly obtained with user authorization.

[0059] In some embodiments, for each link to be processed, the classification tag includes any of the following:

[0060] a) First marker: The first marker indicates that the content of the link to be processed can be accessed normally.

[0061] "Able to access normally" means that when a user clicks the link to be processed, they are successfully redirected to the target page, and the page content loads and renders normally without any network errors or page display failures. This indicates that the network communication for the link to be processed is normal, and the link to be processed is marked as the first flag in the access log information.

[0062] b) The second marker is used to indicate network communication errors of the target type; the target type is used to identify dead links.

[0063] Not all network communication errors are caused by dead links. For example, when a user switches from a WiFi network to a mobile network, a network communication error may occur, but this is not caused by a dead link, so there is no need to adjust the recommended order of such network links. Therefore, in this embodiment of the disclosure, the target type of dead links is mainly marked with a second label.

[0064] The target types include, for example, network layer errors such as -105, -101, and -102, HTTP errors such as 40x and 50x, and response timeouts, as well as rendering errors.

[0065] c) The third marker is used to indicate other cases, which include cases other than those marked by the first and second markers.

[0066] The third tag is used to mark situations where the network resources corresponding to the pending link fail to load or render due to reasons other than dead links. For example, it can be the situation where the user closes the webpage before the page is rendered, or it can include network communication errors other than the aforementioned target types.

[0067] In practice, for ease of recording and transmission, the first flag can be 1, the second flag can be 0, and the third flag can be -1.

[0068] In this embodiment, the first marker is used to record normally accessible links; the second marker focuses on identifying dead links; and the third marker focuses on marking other abnormal situations. These three types of markers enable the server to promptly detect potential dead links, improving the efficiency of network communication error identification and handling. This allows for more accurate optimization of the recommended ranking of target links, thereby enhancing the user experience.

[0069] S302, based on clustering rules, performs clustering analysis on multiple unprocessed links in the log information of multiple terminal devices to obtain the target link set.

[0070] In some embodiments, clustering rules may include clustering based on the domain names of the links to be processed. For example, domain names are identified from the links (URLs) to be processed, and URLs of the same domain name are then grouped into the same set to obtain multiple clusters. It can be understood that each cluster obtained from the clustering analysis in this disclosure embodiment can serve as a target link set for executing the network link processing method provided in this disclosure embodiment. Clustering analysis based on domain names allows for a statistical understanding of whether the processing logic of each communication operation object for the corresponding domain name is normal, thereby facilitating personalized optimization of potential dead links for different communication operation objects.

[0071] Of course, in other embodiments, clustering analysis can also be performed based on domain names according to different functional modules or language versions under the same domain name. In specific implementation, the clustering rules can be determined according to the actual situation, and this disclosure does not limit them.

[0072] S303, based on the classification tag of at least one target link in the target link set and the identifier of the communication operation object, determine the access success rate and the number of access records for each communication operation object in the target link set to obtain statistical records.

[0073] During implementation, if there are multiple communication operation objects, the access success rate and the number of access records for each communication operation object should be determined.

[0074] The number of access records represents the number of times all target links in the target link set were accessed within the statistical period.

[0075] For any communication operation object, its corresponding access success rate is calculated based on the following formula (1):

[0076] D= [A / (Z - C) ]* 100% (1)

[0077] Where D is the success rate of the communication operation object accessing the target link set, A is the number of access records with the first tag under the communication operation object, Z is the total number of access records under the communication operation object, and C is the number of access records with the third tag under the communication operation object.

[0078] Since the third-marked access failure is not due to a potential dead link, in order to improve the accuracy of identifying potential dead links, the number of access records marked with the third mark (C) is deleted from the total access records (Z), resulting in the number of successful access records and the number of access records due to potential dead links.

[0079] In the presence of multiple communication operation objects, each communication operation object obtains its own access success rate based on equation (1). This access success rate is statistically based on potential dead links, which can reflect the potential dead link situation of each communication operation object for the same target link.

[0080] In this embodiment, log information within a statistical period is obtained from multiple terminal devices. This information includes access records, each of which details the pending link, its corresponding category tag, and the communication operation object identifier. Combining the statistical period (time dimension) with the communication operation object constructs two statistical dimensions, enabling more accurate statistics on network communication errors for different communication operation objects across different statistical periods. This allows for dynamic monitoring of potential dead links for different communication operation objects, facilitating reasonable optimization of the recommended ranking of potential dead links and improving user experience.

[0081] In some embodiments, adjusting the recommendation ranking of target links within the target link set in the corresponding resource set to be recommended, based on statistical records, can be implemented as follows:

[0082] For each resource ranking stage of the target link, obtain the corresponding adjustment rules; based on the adjustment rules and statistical records, adjust the recommendation ranking of the target link in the corresponding set of resources to be recommended.

[0083] The resource ranking stage includes both offline and online resource ranking. In a recommendation system, the offline resource ranking stage primarily refers to the process of sorting a large number of candidate web links in an offline environment. The aim is to identify potentially interesting content based on public interests and behavioral history, and arrange them in a more reasonable order for presentation to users in the subsequent online resource recommendation stage.

[0084] The online resource sorting stage is mainly used to respond to users' resource requests by arranging and combining resources according to certain rules or algorithms, so that users can obtain the information they need more efficiently and conveniently.

[0085] The specific methods for adjusting the recommendation ranking in these two stages will be explained later.

[0086] In this embodiment, the target link can be sorted and optimized according to its specific sorting stage. This fully considers the characteristics and needs of the target link at different stages, making the target link more targeted and phased in the corresponding set of recommended resources. The recommendation sorting can be optimized from different stages, improving the quality of the recommended resources finally presented to the user.

[0087] In some embodiments, the recommendation ranking of the target link in the corresponding set of resources to be recommended is adjusted based on adjustment rules and statistical records, such as... Figure 4 As shown, it can be implemented as follows:

[0088] S401, when the resource sorting stage is the offline resource sorting stage, for each communication operation object, based on the adjustment rules of the offline resource sorting stage, determine the first access success rate of the target link set within at least one statistical period in the statistical records, and the first confidence level of the first access success rate.

[0089] In a statistical period, the first access success rate is the access success rate recorded within that statistical period.

[0090] When multiple statistical periods are included, the first access success rate is the statistical value of the access success rates of these multiple statistical periods, such as the mean or a weighted value. The weighting coefficient can be determined based on the time decay factor so that the closer to the current time, the higher the weight. Alternatively, access records from multiple statistical periods can be aggregated, and then the access success rate can be recalculated based on formula (1) to obtain the first access success rate.

[0091] The first confidence level is determined based on the number of access records in the at least one statistical period. The higher the number of access records, the higher the first confidence level of the first access success rate.

[0092] S402, based on the first access success rate of the target link set within at least one statistical period for each communication operation object, and the first confidence level of the first access success rate, adjust the recommendation ranking of the target links in the corresponding set of resources to be recommended.

[0093] The first access success rate can identify whether a link is a potential dead link, and the first confidence score is used to indicate the credibility of the identification result.

[0094] In this embodiment of the disclosure, through offline analysis, a first access success rate and a first confidence level of the target link set within at least one statistical period are determined for each communication operation object. Based on these data, the recommendation ranking is adjusted. This can optimize the recommendation order of network resources for target links during the offline resource ranking stage, so as to provide further optimized resources for the online ranking stage, improve the resource quality during the online resource ranking stage, and enable users to access high-quality links more efficiently, thereby improving the user experience.

[0095] In some embodiments, the recommendation ranking of the target links in the corresponding set of resources to be recommended is adjusted based on the first access success rate of the target link set within at least one statistical period of each communication operation object and the first confidence level of the first access success rate. This can be implemented as follows: if each communication operation object meets the following first condition, the target links of the target link set are deleted from the corresponding set of resources to be recommended: the first condition includes: the first access success rate is lower than the first success rate threshold within multiple consecutive statistical periods, and the first confidence level is higher than the first confidence threshold within multiple consecutive statistical periods.

[0096] For example, the first success rate threshold can be set to 10%, and the first confidence threshold can be set to 50. If the first access success rate of each communication operation object for the target link set is less than 10% in each statistical period, and the first confidence score is greater than 50 in each statistical period, then it is determined to be a stable dead link and can be deleted from the resource set to be recommended. The aforementioned specific values ​​are for illustrative purposes only and are not limited herein.

[0097] In this embodiment of the disclosure, if the access success rate is lower than the first success rate threshold and the first confidence level is higher than the first confidence level within multiple consecutive statistical periods, stable dead links with a long-term low access success rate can be identified and removed from the recommended resource set, so as to minimize the proportion of low-quality links in the recommended resources.

[0098] In addition to identifying the aforementioned set of dead links, this embodiment of the disclosure can also adjust the recommended ranking of some unstable dead links based on their failure status. Specifically, in some embodiments, based on the first access success rate of the target link set within at least one statistical period for each communication operation object, and the first confidence level of the first access success rate, the recommended ranking of the target links in the corresponding set of resources to be recommended can be adjusted. This can be implemented as follows: if each communication operation object meets the following second condition, the ranking of the target links in the target link set is adjusted from the first target ranking position to after the first target ranking position.

[0099] The second condition includes: the first access success rate is less than the second success rate threshold in at least one statistical period, and the fluctuation value of the first access success rate is greater than the preset fluctuation value.

[0100] For example, the second success rate threshold can be 30%, and the preset fluctuation value is used to indicate that the first access success rate fluctuates significantly over multiple consecutive statistical periods. In implementation, the method for determining the fluctuation value and the preset fluctuation threshold can be determined based on the actual situation. For instance, the difference between the first access success rates of different communication operators for the target link set within the same statistical period can be determined. This difference can be expressed using the absolute value of the difference, variance, etc. If this difference is greater than the preset fluctuation value, then the first access success rate is determined to have significant fluctuations.

[0101] For example, the degree of difference in the first access success rate of the same communication operator for the target link set within multiple consecutive statistical periods can be determined. If the degree of difference is greater than a preset fluctuation value, it is determined that the access success rate of the communication operator for the target link set fluctuates significantly.

[0102] The first target ranking position can be ranked 20th.

[0103] Assuming that the first access success rate of the target link is less than 20% for each telecommunications operator in the i-th statistical period, and the first access success rate fluctuates significantly, it may be due to network or other reasons causing abnormal access to the target link set. If each telecommunications operator meets the aforementioned conditions, the target link can be adjusted to a position after 20 to maintain its exposure. Its subsequent performance should be tracked simultaneously to dynamically adjust its recommendation ranking. The aforementioned specific thresholds are illustrative and are not limited herein.

[0104] In this embodiment of the disclosure, when the first access success rate of the target link set is lower than the second success rate threshold and fluctuates greatly, the sorting position of the target links included in the set is shifted to the back, so as to enable users to access more stable and reliable resources first, and retain the exposure opportunity of potential dead links, thereby improving the user experience.

[0105] In some embodiments, as described above, the exposure rate of potentially dead links can be retained, and their subsequent performance can be tracked simultaneously to dynamically adjust their recommendation ranking. Accordingly, after adjusting the ranking of the target link from the corresponding set of recommended resources to the first target ranking position based on the aforementioned method, it can also be implemented as follows: tracking the fourth access success rate and the fourth confidence level of the target link set; if the fourth access success rate is greater than the fourth success rate threshold and the fourth confidence level is greater than the third confidence level threshold, then stopping the adjustment of the target link set from the corresponding set of recommended resources to the first target ranking position.

[0106] During implementation, after adjusting it to the first target ranking position, its low exposure rate is retained. At the same time, based on the aforementioned log information, the target link set to which the target link belongs is tracked, and its fourth access success rate and fourth confidence level are recorded for at least one statistical period in the later period. If the fourth access success rate is greater than the fourth success rate threshold and the fourth confidence level is greater than the third confidence level threshold, the target link can be restored to its previous ranking position, or it can be adjusted to a position before the first target ranking position.

[0107] In this embodiment, by continuously tracking the target link set, the system can promptly detect when the target link set has returned to normal based on the fourth access success rate and the fourth confidence level. This allows for the cessation of adjustments to the target link set beyond its initial ranking position, reducing over-adjustments caused by short-term fluctuations or temporary issues. Adjustments are only stopped when the links are truly stable and reliable. This reduces the impact of misjudgments on the recommendation ranking, improving its stability and reliability.

[0108] In some embodiments, during the online resource ranking phase, adjusting the recommendation ranking of target links in the corresponding set of resources to be recommended based on adjustment rules and statistical records can be implemented as follows:

[0109] Step A1, the online resource sorting stage, responds to the target object's access request for the target link and parses the target communication operation object to which the access request belongs.

[0110] The online resource ranking stage refers to dynamically analyzing the latest data indicators by responding to user access requests in real time, and adjusting the ranking of resources to be recommended accordingly to optimize the quality of resources available to users.

[0111] Step A2: Based on the adjustment rules corresponding to the online resource sorting stage, analyze the second access success rate and second confidence level of the target communication operation object in the latest statistical sub-period in the statistical records; the time length of the statistical sub-period is less than or equal to the statistical period, and the statistical sub-period is determined based on the request time of the access request.

[0112] The calculation methods for the second access success rate and the second confidence level are the same as those described above, except that the statistical time period has changed, which will not be elaborated here.

[0113] In some embodiments, if the time window of the statistical sub-period length where the access request time ends, the sorting result of the target link determined based on the statistical sub-period is determined to be invalid.

[0114] During implementation, the length of the time window for the statistical sub-period can be defined based on business performance. For example, based on user network activity patterns, user traffic is typically lower on weekdays, the network is relatively less congested, and the possibility of potential dead links is also lower, resulting in a higher access success rate for the corresponding statistical period. Conversely, on weekends, the network is busy, and the access success rate may be lower for the corresponding weekend statistical period. Therefore, a sliding window can be set for weekdays as one statistical sub-period, and another sliding window can be set for weekends to obtain the statistical sub-period for weekends.

[0115] Of course, the length of the statistical sub-period can be flexibly set according to the access success rate of each time period. For example, if the network is relatively busy from 6 pm to 9 pm, this time interval can be set as a statistical sub-period, with the time window of this statistical sub-period being 6 pm to 9 pm. If the request time of each access request falls within the time window of the statistical sub-period, the recommendation ranking within that statistical sub-period is adjusted based on the access success rate and confidence level. When the statistical sub-period ends, the failure handling mechanism based on that statistical sub-period is automatically triggered, marking the ranking results based on that statistical sub-period as invalid. Then, the recommendation ranking is re-optimized based on the statistical sub-period of the next sliding window.

[0116] In this embodiment of the disclosure, the timeliness management mechanism ensures a close correlation between the ranking results and the specific circumstances of the statistical sub-period, enabling flexible adjustment of the recommended ranking and improving the rationality and accuracy of the recommended ranking.

[0117] Step A3: If the second access success rate is less than the third success rate threshold and the second confidence level is greater than the second confidence level threshold in the latest statistical sub-period, then the ranking position of the target link in the resource set to be recommended is adjusted to after the second target ranking position.

[0118] For example, if the second confidence threshold is 30 access records and the third success rate threshold is 50%, and the second access success rate in the latest statistical sub-period calculated based on the aforementioned method is 30%, and the second confidence level is 50, then the target link needs to be adjusted to a position after the second target sorting position.

[0119] In this embodiment of the disclosure, during the online resource ranking stage, the recommended ranking of target links in the target connection set can be dynamically adjusted in real time based on this method, so as to recommend reasonable network links to users according to the situation.

[0120] In some embodiments, if the second access success rate and second confidence level within the latest statistical sub-period are obtained based on the aforementioned step A2, the implementation can also be as follows:

[0121] Step B1: If the second access success rate in the latest statistical sub-period is less than the third success rate threshold and the second confidence level is not greater than the second confidence level threshold, summarize the third access success rate and third confidence level of the target communication operation object in multiple consecutive statistical sub-periods, including the latest statistical sub-period.

[0122] In other words, if the second confidence level in the latest statistical period is not greater than the second confidence level threshold, it means that more data is needed to verify the credibility of the second access success rate in order to dynamically adjust the recommendation ranking.

[0123] In implementation, the third access success rate and third confidence level of the target communication operation object are summarized over multiple consecutive statistical sub-periods, including the latest statistical sub-period. This can be implemented as follows: Based on the latest statistical sub-period, obtain k consecutive statistical sub-periods, including that latest statistical sub-period. For example, 6 PM to 9 PM each day is considered a statistical sub-period, then 6 PM to 9 PM on Monday, Tuesday, and Wednesday constitutes three consecutive statistical sub-periods of that period. Obtain the third access success rate and third confidence level of the target communication operation object corresponding to multiple statistical sub-periods, where k is a positive integer greater than or equal to 1.

[0124] During implementation, access records from multiple statistical sub-periods are continuously aggregated, and then the third access success rate and third confidence level of the target communication operation object can be obtained by calculation based on Equation (1).

[0125] Step B2: If the third access success rate is less than the third success rate threshold and the third confidence level is greater than the second confidence level threshold, adjust the ranking position of the target link in the resource set to be recommended to after the second target ranking position.

[0126] For example, if the second confidence threshold is 30, the third success rate threshold is 50%, the third access success rate calculated based on the above method is 35%, and the third confidence is 40, then the target link needs to be adjusted to a position after the second target sorting position.

[0127] In this embodiment, if the second access success rate in the latest statistical sub-period is lower than the third success rate threshold, but the second confidence level is not greater than the second confidence level threshold, it is necessary to aggregate the third access success rate and third confidence level over multiple consecutive statistical sub-periods. This allows for verification of the credibility of the access success rate of the target link set using more data. When the third access success rate is lower than the third success rate threshold and the third confidence level is higher than the second confidence level threshold, the ranking position of the target link is adjusted to a later position. This mechanism reduces misjudgments caused by the randomness or instability of data in a single sub-period. By comprehensively evaluating data from multiple consecutive statistical sub-periods, a more comprehensive understanding of the overall performance of the target link set over a period of time can be obtained, improving the reliability and accuracy of decision-making, enhancing the stability and reliability of link quality assessment, reducing adjustment errors caused by short-term fluctuations or data anomalies, and improving the quality of the recommendation ranking of the recommended resource set.

[0128] In some embodiments, the above thresholds can be dynamically determined as needed. For example, it can be implemented by dynamically optimizing at least one of the following based on A / B testing: a first success rate threshold, a second success rate threshold, a third success rate threshold, a fourth success rate threshold, a first confidence threshold, a second confidence threshold, a third confidence threshold, and a time window of a statistical sub-cycle.

[0129] During implementation, multiple A / B experiments can be conducted with different thresholds to optimize the multiple thresholds, thereby further improving the matching degree between each threshold and the business scenario and improving the accuracy of the adjusted recommendation ranking.

[0130] Based on the same technical concept, this disclosure also proposes a network link processing device 500, such as... Figure 5 As shown, it includes:

[0131] The acquisition module 501 is used to acquire statistical records of the target link set within at least one statistical period; the statistical records are used to record network communication error conditions of multiple communication operation objects.

[0132] The adjustment module 502 is used to adjust the recommendation order of the target links in the target link set in the corresponding resource set to be recommended based on the statistical records.

[0133] In some embodiments, the acquisition module includes:

[0134] The first acquisition unit is used to acquire log information generated by multiple terminal devices within the statistical period. The log information includes multiple access records. Each access record includes: a link to be processed, a classification tag corresponding to the link to be processed, and a communication operation object identifier. The classification tag is used to indicate the network communication status corresponding to the link to be processed.

[0135] The clustering unit is used to perform clustering analysis on multiple unprocessed links in the log information of the multiple terminal devices based on clustering rules to obtain the target link set;

[0136] The first determining unit is used to determine the access success rate and the number of access records of each communication operation object in the target link set based on the classification tag and communication operation object identifier of at least one target link in the target link set, so as to obtain the statistical record.

[0137] In some embodiments, the classification tag for each link to be processed includes any of the following:

[0138] A first marker, which indicates that the content of the link to be processed can be accessed normally;

[0139] A second marker is used to indicate a network communication error of a target type; the target type is used to identify dead links.

[0140] A third marker is used to indicate other cases, which include cases other than the first and second markers.

[0141] In some embodiments, the adjustment module includes:

[0142] The second acquisition unit is used to acquire the adjustment rules corresponding to the resource sorting stage in which the target link is located.

[0143] The adjustment unit is used to adjust the recommendation ranking of the target link in the corresponding set of resources to be recommended based on the adjustment rules and the statistical records.

[0144] In some embodiments, the adjustment unit is configured to:

[0145] In the case that the resource sorting phase is an offline resource sorting phase, the following is performed for each communication operation object:

[0146] Based on the adjustment rules of the offline resource sorting stage, determine the first access success rate of the target link set within at least one statistical period in the statistical records, and the first confidence level of the first access success rate.

[0147] Based on the first access success rate of the target link set within at least one statistical period for each communication operation object, and the first confidence level of the first access success rate, the recommendation ranking of the target links in the target link set in the corresponding set of resources to be recommended is adjusted.

[0148] In some embodiments, the adjustment unit is specifically used for:

[0149] If all communication operators meet the following first condition, the target links in the target link set will be removed from the corresponding resource set to be recommended:

[0150] The first condition includes: the first access success rate is lower than the first success rate threshold for multiple consecutive statistical periods, and the first confidence level is higher than the first confidence threshold for multiple consecutive statistical periods.

[0151] In some embodiments, the adjustment unit is specifically used for:

[0152] If each communication operation object meets the following second condition, the target links in the target link set will be adjusted from the corresponding resource set to be recommended to the first target sorting position;

[0153] The second condition includes: the first access success rate is less than the second success rate threshold in at least one statistical period, and the fluctuation value of the first access success rate is greater than the preset fluctuation value.

[0154] In some embodiments, the adjustment unit is configured to:

[0155] During the online resource sorting phase, in response to the target object's access request for the target link, the target communication operation object to which the access request belongs is parsed.

[0156] Based on the adjustment rules corresponding to the online resource sorting stage, the second access success rate and second confidence level of the target communication operation object in the latest statistical sub-period are analyzed in the statistical records; the time length of the statistical sub-period is less than or equal to the statistical period, and the statistical sub-period is determined based on the request time of the access request;

[0157] If the second access success rate in the latest statistical sub-period is less than the third success rate threshold and the second confidence level is greater than the second confidence threshold, the ranking position of the target link in the set of resources to be recommended will be adjusted to the second target ranking position.

[0158] In some embodiments, an update module is also included, for:

[0159] If the second access success rate in the latest statistical sub-period is less than the third success rate threshold, and the second confidence level is not greater than the second confidence level threshold, the third access success rate and third confidence level of the target communication operation object in multiple consecutive statistical sub-periods, including the latest statistical sub-period, are aggregated.

[0160] If the third access success rate is less than the third success rate threshold and the third confidence level is greater than the second confidence level threshold, the ranking position of the target link in the set of resources to be recommended will be adjusted to after the second target ranking position.

[0161] In some embodiments, a determining module is further included, configured to:

[0162] If the time window containing the statistical sub-period length of the access request time ends, the sorting result of the target link determined based on the statistical sub-period is determined to be invalid.

[0163] In some embodiments, a tracking module is also included for:

[0164] Track the fourth access success rate and fourth confidence level of the target link set;

[0165] If the fourth access success rate is greater than the fourth success rate threshold and the fourth confidence level is greater than the third confidence level threshold, then stop adjusting the ranking of the target links in the target link set from the corresponding resource set to be recommended to the first target ranking position.

[0166] The specific functions and examples of each module and submodule of the apparatus in this disclosure can be found in the relevant descriptions of the corresponding steps in the above method embodiments, and will not be repeated here.

[0167] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0168] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0169] Figure 6 A schematic block diagram of an example electronic device 600 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0170] like Figure 6As shown, device 600 includes a computing unit 601, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 602 or a computer program loaded from storage unit 608 into random access memory (RAM) 603. RAM 603 may also store various programs and data required for the operation of device 600. The computing unit 601, ROM 602, and RAM 603 are interconnected via bus 604. Input / output (I / O) interface 605 is also connected to bus 604.

[0171] Multiple components in device 600 are connected to I / O interface 605, including: input unit 606, such as keyboard, mouse, etc.; output unit 607, such as various types of monitors, speakers, etc.; storage unit 608, such as disk, optical disk, etc.; and communication unit 609, such as network card, modem, wireless transceiver, etc. Communication unit 609 allows device 600 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0172] The computing unit 601 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 performs the various methods and processes described above, such as network link processing methods. For example, in some embodiments, the network link processing methods may be implemented as computer software programs tangibly contained in a machine-readable medium, such as storage unit 608. In some embodiments, part or all of the computer program may be loaded and / or installed on device 600 via ROM 602 and / or communication unit 609. When the computer program is loaded into RAM 603 and executed by the computing unit 601, one or more steps of the network link processing methods described above may be performed. Alternatively, in other embodiments, the computing unit 601 may be configured to perform network link processing methods by any other suitable means (e.g., by means of firmware).

[0173] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0174] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0175] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0176] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0177] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0178] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0179] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0180] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for processing network links, comprising: Retrieve statistical records of the target link set for at least one statistical period; The statistical records are used to record network communication errors of multiple communication operation objects; Based on the statistical records, the recommendation ranking of the target links in the target link set is adjusted in the corresponding resource set to be recommended.

2. The method according to claim 1, wherein, For each statistical period, generate statistical records for the target link set within the statistical period, including: The system acquires log information generated by multiple terminal devices within the statistical period. The log information includes multiple access records. Each access record includes: a link to be processed, a classification tag corresponding to the link to be processed, and a communication operation object identifier. The classification tag is used to indicate the network communication status corresponding to the link to be processed. Based on clustering rules, clustering analysis is performed on multiple unprocessed links in the log information of the multiple terminal devices to obtain the target link set. Based on the classification tag and communication operation object identifier of at least one target link in the target link set, the access success rate and the number of access records of each communication operation object in the target link set are determined to obtain the statistical records.

3. The method according to claim 2, wherein, For each link to be processed, the classification tag includes any of the following: A first marker, which indicates that the content of the link to be processed can be accessed normally; A second marker is used to indicate a network communication error of a target type; the target type is used to identify dead links. A third marker is used to indicate other cases, which include cases other than the first and second markers.

4. The method according to any one of claims 1-3, wherein, The step of adjusting the recommendation ranking of target links within the target link set in the corresponding resource set to be recommended, based on the statistical records, includes: For the resource sorting stage in which the target link is located, obtain the adjustment rules corresponding to the resource sorting stage; Based on the adjustment rules and the statistical records, the recommendation ranking of the target link in the corresponding set of resources to be recommended is adjusted.

5. The method according to claim 4, wherein, The step of adjusting the recommendation ranking of the target link in the corresponding set of resources to be recommended based on the adjustment rules and the statistical records includes: When the resource sorting phase is an offline resource sorting phase, the following is performed for each communication operation object: based on the adjustment rules of the offline resource sorting phase, determine the first access success rate of the target link set within at least one statistical period in the statistical records, and the first confidence level of the first access success rate; Based on the first access success rate of the target link set within at least one statistical period for each communication operation object, and the first confidence level of the first access success rate, the recommendation ranking of the target links in the target link set in the corresponding set of resources to be recommended is adjusted.

6. The method according to claim 5, wherein, The step of adjusting the recommendation ranking of the target links in the corresponding set of resources to be recommended based on the first access success rate of the target link set within at least one statistical period for each communication operation object, and the first confidence level of the first access success rate, includes: If all communication operators meet the following first condition, the target links in the target link set will be removed from the corresponding resource set to be recommended: The first condition includes: the first access success rate is lower than the first success rate threshold for multiple consecutive statistical periods, and the first confidence level is higher than the first confidence threshold for multiple consecutive statistical periods.

7. The method according to claim 5, wherein, The step of adjusting the recommendation ranking of the target links in the corresponding set of resources to be recommended based on the first access success rate of the target link set within at least one statistical period for each communication operation object, and the first confidence level of the first access success rate, includes: If each communication operation object meets the following second condition, the target links in the target link set will be adjusted from the corresponding resource set to be recommended to the first target sorting position; The second condition includes: the first access success rate is less than the second success rate threshold in at least one statistical period, and the fluctuation value of the first access success rate is greater than the preset fluctuation value.

8. The method according to claim 4, wherein, The step of adjusting the recommendation ranking of the target link in the corresponding set of resources to be recommended based on the adjustment rules and the statistical records includes: During the online resource sorting phase, in response to the target object's access request for the target link, the target communication operation object to which the access request belongs is parsed. Based on the adjustment rules corresponding to the online resource sorting stage, the second access success rate and second confidence level of the target communication operation object in the latest statistical sub-period are analyzed in the statistical records; the time length of the statistical sub-period is less than or equal to the statistical period, and the statistical sub-period is determined based on the request time of the access request; If the second access success rate in the latest statistical sub-period is less than the third success rate threshold and the second confidence level is greater than the second confidence threshold, the ranking position of the target link in the set of resources to be recommended will be adjusted to the second target ranking position.

9. The method according to claim 8, further comprising: If the second access success rate in the latest statistical sub-period is less than the third success rate threshold, and the second confidence level is not greater than the second confidence level threshold, the third access success rate and third confidence level of the target communication operation object in multiple consecutive statistical sub-periods, including the latest statistical sub-period, are aggregated. If the third access success rate is less than the third success rate threshold and the third confidence level is greater than the second confidence level threshold, the ranking position of the target link in the set of resources to be recommended will be adjusted to after the second target ranking position.

10. The method according to claim 8 or 9, further comprising: If the time window containing the statistical sub-period length of the access request time ends, the sorting result of the target link determined based on the statistical sub-period is determined to be invalid.

11. The method of claim 7, further comprising: Track the fourth access success rate and fourth confidence level of the target link set; If the fourth access success rate is greater than the fourth success rate threshold and the fourth confidence level is greater than the third confidence level threshold, then stop adjusting the ranking of the target links in the target link set from the corresponding resource set to be recommended to the first target ranking position.

12. A network link processing apparatus, comprising: The acquisition module is used to acquire statistical records of the target link set within at least one statistical period; The statistical records are used to record network communication errors of multiple communication operation objects; The adjustment module is used to adjust the recommendation ranking of target links in the target link set within the corresponding resource set to be recommended, based on the statistical records.

13. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-11.

14. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-11.

15. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-11.

Images