Page push method, device, electronic device and storage medium

Abstract

The present disclosure proposes a page push method, device, electronic device and storage medium, which relate to the field of artificial intelligence technology, specifically to the technical fields of natural language processing, intelligent search, deep learning, etc. The solution is: when a new first page is detected in the page set to be pushed, the first page is identified as an abnormal page; when the first page is identified as abnormal, the domain name of the first page is used as the abnormal domain name, and the second page subsequently added to the page set is monitored to select related pages with abnormal domain names from the second page; the related pages are identified as abnormal pages; according to the proportion of pages identified as abnormal pages in the related pages, the push probability of each page under the abnormal domain name in the page set is adjusted, thereby achieving the management and control of pages with lower page quality in the page set to be pushed, so that pages with higher page quality can be pushed, thereby improving the user experience.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, specifically to technical fields such as natural language processing, intelligent search, and deep learning, and more particularly to a page push method, device, electronic device, and storage medium. Background Art

[0002] With the continuous advancement of commercialization, the quality of pages pushed on the Internet (such as advertising pages) is uneven. Pushing some low-quality pages may cause the web pages to be unable to open, false comments on the web pages, etc.; in addition, pushing low-quality pages will also take up the space of high-quality pages, affecting the user experience, and in serious cases may cause public opinion. Therefore, how to push pages is very important. Summary of the Invention

[0003] The present disclosure provides a method, device, electronic device, and storage medium for page push.

[0004] According to one aspect of the present disclosure, a page push method is provided, comprising: upon monitoring a newly added first page in a set of pages to be pushed, identifying the first page as an abnormal page; upon identifying that the first page is abnormal, using the domain name of the first page as an abnormal domain name, monitoring a subsequently added second page in the page set to select related pages with the abnormal domain name from the second pages; identifying the related pages as abnormal pages; and adjusting the push probability of each page under the abnormal domain name in the page set according to the proportion of abnormal pages identified in the related pages.

[0005] According to another aspect of the present disclosure, a page pushing device is provided, including: an identification module for detecting that a first page is newly added to a set of pages to be pushed, and identifying the first page as an abnormal page; a monitoring module for detecting that the first page is abnormal, and using the domain name of the first page as an abnormal domain name to monitor a second page subsequently added to the set of pages, so as to select related pages with the abnormal domain name from the second pages; the identification module is further used to identify the related pages as abnormal pages; and an adjustment module is used to adjust the push probability of each page under the abnormal domain name in the set of pages according to the proportion of the related pages identified as abnormal pages.

[0006] According to another aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the page push method described in the embodiment of the first aspect of the present disclosure.

[0007] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the page pushing method described in the embodiment of the first aspect of the present disclosure.

[0008] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, which, when executed by a processor, implements the page pushing method described in the embodiment of the first aspect of the present disclosure.

[0009] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The accompanying drawings are used to better understand the present invention and do not constitute a limitation of the present invention.

[0011] Figure 1 This is a flow chart of the page push method provided in the first embodiment of the present disclosure;

[0012] Figure 2 This is a flow chart of the page push method provided in the second embodiment of the present disclosure;

[0013] Figure 3 This is a flow chart of the page push method provided in the third embodiment of the present disclosure;

[0014] Figure 4 This is a flow chart of the page push method provided in the fourth embodiment of the present disclosure;

[0015] Figure 5 This is a flow chart of the page push method provided in the fifth embodiment of the present disclosure;

[0016] Figure 6 This is a flow chart of the page push method provided in the sixth embodiment of the present disclosure;

[0017] Figure 7 This is a flow chart of the page push method provided in the seventh embodiment of the present disclosure;

[0018] Figure 8A schematic diagram of abnormal page identification provided by an embodiment of the present disclosure;

[0019] Figure 9 This is a structural diagram of the page pushing device provided in the eighth embodiment of the present disclosure;

[0020] Figure 10 It is a block diagram of an electronic device used to implement the page push method of the embodiment of the present disclosure. DETAILED DESCRIPTION

[0021] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0022] At present, the quality of pages pushed on the Internet (such as advertising pages) is uneven, and a large number of pages (landing pages) have some quality problems, mainly reflected in excessive marketing of products, inability to open web pages, false comments on web pages, etc.; low-quality pages will not only squeeze the space of high-quality pages and affect the user experience, but in serious cases may even cause public opinion, so it is very important to identify the quality of page content; similarly, in order to create a good product ecology, the identification of low-quality advertisements is imperative.

[0023] In related technologies, manual annotation is often used to review page materials submitted by the page owner, and the materials are rejected when there are certain risks. The main risks reviewed are more serious risks, and risks that affect user experience, such as dead links and abnormal pages, are not controlled.

[0024] In response to the above problems, the present disclosure proposes a page push method, device, electronic device and storage medium.

[0025] The following describes the page push method, device, electronic device, and storage medium according to the embodiments of the present disclosure with reference to the accompanying drawings.

[0026] Figure 1 This is a flow chart of the page push method provided in the first embodiment of the present disclosure.

[0027] The embodiment of the present disclosure is illustrated by taking the page pushing method configured in a page pushing device as an example. The page pushing device can be applied to any electronic device so that the electronic device can perform a page pushing function.

[0028] Among them, the electronic device can be any device with computing capabilities, such as a personal computer (PC), a mobile terminal, etc. The mobile terminal can be, for example, a mobile phone, a tablet computer, a personal digital assistant, a wearable device, and other hardware devices with various operating systems, touch screens and / or display screens.

[0029] like Figure 1 As shown, the page push method may include the following steps:

[0030] Step 101: When it is detected that a first page is newly added to a set of pages to be pushed, abnormal page identification is performed on the first page.

[0031] In an embodiment of the present disclosure, the page set to be pushed may include multiple pages to be pushed. When a first page is newly added to the page set to be pushed, abnormal page identification may be performed on the first page to determine whether there is an abnormality in the first page.

[0032] Step 102: When an abnormality is identified in the first page, the domain name of the first page is used as an abnormal domain name, and the second page subsequently added to the page set is monitored to select related pages with abnormal domain names from the second pages.

[0033] In an embodiment of the present disclosure, when it is identified that there is an abnormality in the first page, the domain name of the first page can be obtained, and the domain name of the first page can be used as the abnormal domain name. Then, the domain name of the second page subsequently added to the page set can be obtained, and the domain name of the second page can be compared with the abnormal domain name to select related pages with abnormal domain names from the second page.

[0034] Step 103: Identify abnormal pages on the relevant pages.

[0035] Then, abnormal page identification is performed on the relevant pages to determine whether there are any abnormalities in the relevant pages.

[0036] Step 104 : According to the proportion of abnormal pages identified in the relevant pages, the push probability of each page under the abnormal domain name in the page set is adjusted.

[0037] In the embodiment of the present disclosure, the push probability of each page under the abnormal domain name in the page set can be adjusted according to the page proportion of abnormal pages identified in the relevant pages, that is, the push of each page under the abnormal domain name in the page set can be controlled.

[0038] In summary, by monitoring the case where a new first page is added to the page set to be pushed, the first page is identified as an abnormal page; when it is identified that the first page is abnormal, the domain name of the first page is used as the abnormal domain name, and the second page subsequently added to the page set is monitored to select related pages with abnormal domain names from the second pages; the related pages are identified as abnormal pages; according to the page ratio of abnormal pages identified in the related pages, the push probability of each page under the abnormal domain name in the page set is adjusted. Therefore, by identifying the abnormal pages under the abnormal domain name and adjusting the push probability of each page under the abnormal domain name in the page set according to the page ratio of abnormal pages under the abnormal domain name, it is possible to manage the pages with lower page quality in the page set to be pushed, so that pages with higher page quality can be pushed, thereby improving the user experience.

[0039] In order to clearly illustrate how the above embodiment adjusts the push probability of each page under the abnormal domain name in the page set according to the page ratio of the abnormal pages identified in the relevant pages, the present disclosure proposes another page push method.

[0040] Figure 2 This is a flow chart of the page push method provided in the second embodiment of the present disclosure.

[0041] like Figure 2 As shown, the page push method may include the following steps:

[0042] Step 201: When a first page is detected to be newly added to the set of pages to be pushed, abnormal page identification is performed on the first page.

[0043] Step 202: When an abnormality is identified in the first page, the domain name of the first page is used as an abnormal domain name, and the second page subsequently added to the page set is monitored to select related pages with abnormal domain names from the second pages.

[0044] Step 203: Identify abnormal pages on the relevant pages.

[0045] Step 204 : When the proportion of abnormal pages identified in the relevant pages is greater than a first ratio threshold, the push probability of each page under the abnormal domain name in the page set is reduced.

[0046] As an example, the page ratio of abnormal pages identified in the relevant pages is compared with the first ratio threshold. When the page ratio of abnormal pages identified in the relevant pages is greater than the first ratio threshold, the push probability of each page under the abnormal domain name in the page set is reduced.

[0047] Step 205 : When the proportion of abnormal pages identified in the relevant pages is less than or equal to the first ratio threshold, the push probability of each page under the abnormal domain name in the page set is restored to the set probability.

[0048] As another example, the page ratio of abnormal pages identified in the relevant pages is compared with the first ratio threshold. When the page ratio of abnormal pages identified in the relevant pages is less than or equal to the first ratio threshold, the push probability of each page under the abnormal domain name in the page set can be restored to the set probability.

[0049] It should be noted that step 204 and step 205 are two parallel execution modes. In actual execution, one is selected. The execution process of steps 201 to 203 can be implemented in any of the embodiments of the present disclosure. The embodiments of the present disclosure do not limit this and will not be described in detail.

[0050] In summary, by identifying that the proportion of abnormal pages in the relevant pages is greater than the first proportion threshold, the push probability of each page under the abnormal domain name in the page set is reduced; when the proportion of abnormal pages in the relevant pages is less than or equal to the first proportion threshold, the push probability of each page under the abnormal domain name in the page set is restored to the set probability. Therefore, by adjusting the push probability of each page under the abnormal domain name in the page set, it is possible to control the pages with lower page quality in the page set to be pushed, so that pages with higher page quality can be pushed, thereby improving the user experience.

[0051] In order to clearly illustrate how, in the above embodiment, the push probability of each page under the abnormal domain name in the page set is reduced when the page proportion of abnormal pages identified in the relevant pages is greater than the first ratio threshold, the present disclosure proposes another page push method.

[0052] Figure 3 This is a flow chart of the page push method provided in the third embodiment of the present disclosure.

[0053] like Figure 3 As shown, the page push method may include the following steps:

[0054] Step 301: When it is detected that a first page is newly added to the set of pages to be pushed, abnormal page identification is performed on the first page.

[0055] Step 302: When an abnormality is identified in the first page, the domain name of the first page is used as an abnormal domain name, and the second page subsequently added to the page set is monitored to select related pages with abnormal domain names from the second pages.

[0056] Step 303: Identify abnormal pages on the relevant pages.

[0057] Step 304 : When the proportion of abnormal pages identified in the relevant pages is greater than a first ratio threshold, determine the pages under the abnormal domain name from the page set.

[0058] In an embodiment of the present disclosure, when the proportion of abnormal pages identified in the relevant pages is greater than a first ratio threshold, each page belonging to the abnormal domain name is determined from the page set.

[0059] Step 305: identify abnormal pages for each page under the abnormal domain name in the page set.

[0060] In an embodiment of the present disclosure, the content of each page under an abnormal domain name in a page set may be identified to determine whether each page is an abnormal page.

[0061] Step 306: Reduce the push probability of abnormal pages under abnormal domain names to a first probability value, and reduce the push probability of non-abnormal pages under abnormal domain names to a second probability value.

[0062] The second probability value is greater than the first probability value.

[0063] Furthermore, the push probability of abnormal pages under abnormal domain names is reduced to a first probability value, and the push probability of non-abnormal pages under abnormal domain names is reduced to a second probability value, wherein the second probability value is greater than the first probability value, that is, the push probability of non-abnormal pages under abnormal domain names is greater than the push probability of abnormal pages under abnormal domain names.

[0064] It should be noted that the execution process of steps 301 to 303 can be implemented in any of the embodiments of the present disclosure, and the embodiments of the present disclosure do not limit this and will not be described in detail.

[0065] In summary, by identifying that the proportion of abnormal pages in the relevant pages is greater than the first proportion threshold, the pages under the abnormal domain name are determined from the page set; the abnormal pages are identified for each page under the abnormal domain name in the page set; the push probability of the abnormal page under the abnormal domain name is reduced to a first probability value, and the push probability of the non-abnormal page under the abnormal domain name is reduced to a second probability value. Thus, the push probability of the abnormal page under the abnormal domain name is reduced to the first probability value, and the push probability of the non-abnormal page under the abnormal domain name is reduced to the second probability value, wherein the second probability value is greater than the first probability value. By adjusting the push probability of each page under the abnormal domain name in the page set, it is possible to manage and control the pages with lower page quality in the page set, so that pages with higher page quality can be pushed, thereby improving the user experience.

[0066] In order to clearly illustrate how, in the above embodiment, the push probability of each page under the abnormal domain name in the page set is reduced when the page proportion of abnormal pages identified in the related pages is greater than the first ratio threshold, the present disclosure proposes another page push method.

[0067] Figure 4 This is a flow chart of the page push method provided in the fourth embodiment of the present disclosure.

[0068] like Figure 4 As shown, the page push method may include the following steps:

[0069] Step 401: When it is detected that a first page is newly added to the set of pages to be pushed, abnormal page identification is performed on the first page.

[0070] Step 402: When an abnormality is identified in the first page, the domain name of the first page is used as an abnormal domain name, and the second page subsequently added to the page set is monitored to select related pages with abnormal domain names from the second page.

[0071] Step 403: Identify abnormal pages on the relevant pages.

[0072] Step 404 : When the proportion of abnormal pages identified in the relevant pages is greater than a first ratio threshold, determine the pages under the abnormal domain name from the page set.

[0073] In an embodiment of the present disclosure, when the proportion of abnormal pages identified in the relevant pages is greater than a first ratio threshold, each page belonging to the abnormal domain name is determined from the page set.

[0074] Step 405: query the publisher of each page under the abnormal domain name in the page set to determine the page belonging to the abnormal publisher.

[0075] Among them, the proportion of abnormal pages in the pages delivered by the abnormal deliverer is greater than the second proportion threshold.

[0076] Furthermore, the publisher of each page under the abnormal domain name in the page collection is queried, and the proportion of abnormal pages in each page is compared with the second ratio threshold, so as to select the pages with the proportion of abnormal pages greater than the second ratio threshold from the pages under the abnormal domain name in the page collection, and the pages with the proportion of abnormal pages greater than the second ratio threshold from the pages under the abnormal domain name in the page collection are regarded as pages of abnormal publishers.

[0077] Step 406: For pages belonging to abnormal publishers, the push probability is reduced to a third probability value; for pages not belonging to abnormal publishers, the push probability is reduced to a fourth probability value.

[0078] The fourth probability value is greater than the third probability value.

[0079] Furthermore, for pages belonging to abnormal publishers, the push probability is reduced to a third probability value, and for pages not belonging to abnormal publishers, the push probability is reduced to a fourth probability value, wherein the fourth probability value is greater than the third probability value.

[0080] It should be noted that the execution process of steps 401 to 403 can be implemented in any of the embodiments of the present disclosure, and the embodiments of the present disclosure do not limit this and will not be described in detail.

[0081] In summary, by identifying that the proportion of abnormal pages in the relevant pages is greater than the first proportion threshold, the pages under the abnormal domain name are determined from the page set; the publisher of each page under the abnormal domain name in the page set is queried to determine the pages belonging to the abnormal publisher; for the pages belonging to the abnormal publisher, the push probability is reduced to the third probability value, and for the pages that do not belong to the abnormal publisher, the push probability is reduced to the fourth probability value. Therefore, by adjusting the push probability of the pages belonging to the abnormal publisher to the third probability value and reducing the push probability of the pages that do not belong to the abnormal publisher to the fourth probability value, the fourth probability value is greater than the third probability value, thereby achieving the management and control of pages with lower page quality in the page set to be pushed, so that pages with higher page quality can be pushed, thereby improving the user experience.

[0082] In order to clearly illustrate how the above embodiment pushes a page to the client, the present disclosure proposes another page pushing method.

[0083] Figure 5 This is a flow chart of the page push method provided in the fifth embodiment of the present disclosure.

[0084] like Figure 5 As shown, the page push method may include the following steps:

[0085] Step 501: When it is detected that a first page is newly added to the set of pages to be pushed, abnormal page identification is performed on the first page.

[0086] Step 502: When an abnormality is identified in the first page, the domain name of the first page is used as an abnormal domain name, and the second page subsequently added to the page set is monitored to select related pages with the abnormal domain name from the second page.

[0087] Step 503: Identify abnormal pages on the relevant pages.

[0088] Step 504 : According to the proportion of abnormal pages identified in the relevant pages, the push probability of each page under the abnormal domain name in the page set is adjusted.

[0089] Step 505 : In response to the trigger word sent by the client, a candidate page matching the trigger word is selected from the page set.

[0090] In an embodiment of the present disclosure, a user may input a trigger word on the client, and the client may send the trigger word to the server. The server responds to the trigger word sent by the client and selects candidate pages whose page content matches the trigger word from a page set.

[0091] Step 506 : sorting the candidate pages according to their push probabilities, and selecting a set number of pages ranked first to push to the client.

[0092] Then, the push probabilities of the candidate pages are sorted from large to small, and according to the arrangement order of the candidate pages, a set number of pages ranked first are pushed to the client.

[0093] It should be noted that the execution process of steps 501 to 504 can be implemented in any of the ways in the embodiments of the present disclosure, and the embodiments of the present disclosure do not limit this and will not be described in detail.

[0094] In summary, by responding to the trigger word sent by the client, candidate pages that match the trigger word are selected from the page set; the candidate pages are sorted according to their push probability, so as to select a set number of pages that are ranked first and pushed to the client. In this way, a set number of pages that match the trigger word and have a high push probability are pushed to the client, so that pages that match the trigger word and have a higher page quality can be pushed to the client.

[0095] In order to clearly illustrate how the above embodiment, when an abnormality is identified in the first page, uses the domain name of the first page as an abnormal domain name, monitors the second page subsequently added to the page collection, and selects related pages with the abnormal domain name from the second page, the present disclosure proposes another page push method.

[0096] Figure 6 This is a flow chart of the page push method provided in the sixth embodiment of the present disclosure.

[0097] like Figure 6 As shown, the page push method may include the following steps:

[0098] Step 601: When it is detected that a first page is newly added to the set of pages to be pushed, abnormal page identification is performed on the first page.

[0099] Step 602 : When it is identified that there are abnormal pages in the first page, the page ratio of the abnormal pages in the first page is determined.

[0100] In an embodiment of the present disclosure, when it is identified that there are abnormal pages in the first page, the abnormal pages in the first page can be identified and compared with the first page to determine the page ratio of the abnormal pages in the first page.

[0101] Step 603: When the proportion of abnormal pages in the first page is greater than a third ratio threshold, the domain name of the first page is regarded as an abnormal domain name.

[0102] Furthermore, the page ratio of abnormal pages in the first page is compared with the third ratio threshold to determine whether the page ratio of abnormal pages in the first page is greater than the third ratio threshold. As an example, when the page ratio of abnormal pages in the first page is greater than the third ratio threshold, the domain name of the first page is regarded as an abnormal domain name.

[0103] As another example, when the proportion of abnormal pages in the first page is less than or equal to the third ratio threshold, the domain name of the first page is regarded as a non-abnormal domain name.

[0104] Step 604: monitor the second pages subsequently added to the page set based on the abnormal domain name to obtain pages with abnormal domain names in the second pages.

[0105] Furthermore, a second page subsequently added to the page set is monitored, and the domain name of the second page is matched with the abnormal domain name, so that the page with the abnormal domain name in the second page can be obtained.

[0106] Step 605: Sample and select relevant pages from the pages with abnormal domain names in the second page.

[0107] As an example, relevant pages may be randomly sampled from pages with abnormal domain names in the second page.

[0108] As an example, pages with abnormal domain names in the second page can be regarded as related pages.

[0109] Step 606: Identify abnormal pages on the relevant pages.

[0110] Step 607: Adjust the push probability of each page under the abnormal domain name in the page set according to the page ratio of the abnormal pages identified in the relevant pages.

[0111] It should be noted that the execution process of step 601 and steps 606 to 607 can be implemented in any way in the embodiments of the present disclosure, and the embodiments of the present disclosure do not limit this and will not be described in detail.

[0112] In summary, by identifying the presence of abnormal pages in the first page, the page ratio of abnormal pages in the first page is determined; when the page ratio of abnormal pages in the first page is greater than the third ratio threshold, the domain name of the first page is used as the abnormal domain name; based on the abnormal domain name, the second page subsequently added to the page set is monitored to obtain pages with abnormal domain names in the second page; from the pages with abnormal domain names in the second page, relevant pages are sampled and selected, thereby accurately determining the abnormal domain name of the first page with the abnormality, and accurately determining the relevant pages with abnormal domain names from the second page.

[0113] In order to clearly illustrate how the above embodiment identifies abnormal pages for the first page, or identifies abnormal pages for related pages, the present disclosure proposes another page push method.

[0114] Figure 7 This is a flow chart of the page push method provided in the seventh embodiment of the present disclosure.

[0115] like Figure 7 As shown, the page push method may include the following steps:

[0116] Step 701 : When it is detected that a first page is newly added to the set of pages to be pushed, feature extraction is performed on the set content of the first page to be identified.

[0117] The setting content includes one or more combinations of page text content, industry information, associated words, attributes of the publisher, and visitor's access behavior information.

[0118] In an embodiment of the present disclosure, when it is detected that a new first page is added to the set of pages to be pushed, feature extraction is performed on the setting content of the first page, where the setting content may include one or more combinations of page text content, industry information, associated words, attributes of the publisher, and visitor's access behavior information.

[0119] In order to accurately extract features of the set content of the first page, as an example, for the first page to be identified, the page text content is extracted; the page text content is segmented to determine the word vector of each word; the word vectors of each word belonging to the same sentence are fused to obtain a sentence vector; and the sentence vector of each sentence in the text part is used as a feature of the page text content.

[0120] That is to say, the text content of the first page to be identified can be extracted to obtain the page text content of the first page. Then, the page text content can be segmented to obtain the word vector of each word. The word vectors of each noun in the same sentence can be fused to obtain the sentence vector. Thus, the sentence vector of each sentence in the text part can be used as the feature of the page text content.

[0121] For example, the word vector of each word is represented as w i , the word vectors of each word in the same sentence are fused, and the resulting sentence vector can be expressed as

[0122] Step 702 : Classify the extracted features using a classification model to determine whether the first page is an abnormal page based on an output of the classification model.

[0123] Furthermore, the classification model can classify each sentence vector to obtain the probability distribution of each category (whether it is abnormal). Based on the probability distribution of each category, the classification model outputs the corresponding predicted category, and based on the predicted category, it can be determined whether the first page is an abnormal page.

[0124] It should be noted that before using the classification model to classify the extracted features, the sentence vector can be mapped into the probability distribution of each category through the softmax in the classification model, and then the cross entropy can be used as the loss function to train the classification model. It can be expressed as the following formula:

[0125]

[0126] Where N is the number of samples, y n The label obtained by marking the nth sample, s n is the sentence vector representation of the nth sample, f is the softmax function, A and B are parameter matrices.

[0127] Step 703: When it is identified that the first page is abnormal, the domain name of the first page is used as an abnormal domain name, and the second page subsequently added to the page set is monitored to select related pages with abnormal domain names from the second pages.

[0128] Step 704 : extracting features from the setting contents of the relevant pages to be identified.

[0129] The setting content includes one or more combinations of page text content, industry information, associated words, attributes of the publisher, and visitor's access behavior information.

[0130] In order to accurately extract features of the set content of relevant pages, as an example, for the relevant pages to be identified, the page text content is extracted; the page text content is segmented to determine the word vector of each word; the word vectors of each word belonging to the same sentence are fused to obtain a sentence vector; and the sentence vector of each sentence in the text part is used as the feature of the page text content.

[0131] Step 705 : Classify the extracted features using a classification model to determine whether the relevant page is an abnormal page based on the output of the classification model.

[0132] Furthermore, the classification model can classify each sentence vector to obtain the probability distribution of each category. Based on the probability distribution of each category, the classification model outputs the corresponding predicted category, and according to the predicted category, it can be determined whether the relevant page is an abnormal page.

[0133] It should be noted that in order to improve the quality of page push, the page text content of each page in the page set is obtained; when the page text content matches the set dead link keyword, the corresponding page is determined to be a dead link page; and the dead link page is deleted from the page set.

[0134] That is to say, it can be determined whether the page text content of each page in the page set matches the set dead link keyword. If the page text content of each page matches the set dead link keyword, the corresponding page is determined to be a dead link page, and the dead link page can be deleted from the page set.

[0135] Step 706 : According to the proportion of abnormal pages identified in the relevant pages, the push probability of each page under the abnormal domain name in the page set is adjusted.

[0136] For example, if Figure 8 As shown, taking the first page or related page as the advertisement landing page as an example, the advertisement landing page can be parsed to obtain the text content of the advertisement page. Abnormal pages and dead links can be identified based on the obtained text content. When the advertisement landing page has any of the above risks, it can be considered that there is a problem with the advertisement landing page, so that the push probability of the advertisement landing page can be controlled.

[0137] It should be noted that the execution process of step 703 and step 706 can be implemented in any way in the embodiments of the present disclosure, and the embodiments of the present disclosure do not limit this and will not be described in detail.

[0138] In summary, feature extraction is performed by setting content for the first page to be identified or related pages; wherein the setting content includes one or more combinations of page text content, industry information, associated words, attributes of the publisher, and visitor access behavior information; a classification model is used to classify the extracted features to determine whether it is an abnormal page based on the output of the classification model. Thus, whether the first page or the related page is an abnormal page is determined by the output of the classification model, thereby improving the accuracy of determining whether the first page or the related page is an abnormal page.

[0139] The page push method of the embodiment of the present invention detects that a new first page is added to the page set to be pushed, and identifies the first page as an abnormal page; when it is identified that the first page is abnormal, the domain name of the first page is used as the abnormal domain name, and the second page subsequently added to the page set is monitored to select related pages with the abnormal domain name from the second page; the related pages are identified as abnormal pages; according to the page ratio of the identified abnormal pages in the related pages, the push probability of each page under the abnormal domain name in the page set is adjusted. Therefore, by identifying the abnormal pages under the abnormal domain name and adjusting the push probability of each page under the abnormal domain name in the page set according to the page ratio of the abnormal pages under the abnormal domain name, it is possible to manage the pages with lower page quality in the page set to be pushed, and push pages with higher page quality, thereby improving the user experience.

[0140] In order to implement the above embodiment, the present disclosure proposes another page pushing device.

[0141] Figure 9 This is a structural diagram of the page pushing device provided in Example 8 of the present disclosure.

[0142] like Figure 9 As shown, the page pushing device 900 includes: an identification module 910 , a monitoring module 920 and an adjustment module 930 .

[0143] Among them, the identification module 910 is used to detect the addition of a new first page to the page set to be pushed and identify the first page as an abnormal page; the monitoring module 920 is used to use the domain name of the first page as an abnormal domain name when it is identified that there is an abnormality in the first page, and monitor the second page subsequently added to the page set to select related pages with the abnormal domain name from the second page; the identification module 910 is also used to identify related pages as abnormal pages; the adjustment module 930 is used to adjust the push probability of each page under the abnormal domain name in the page set according to the page ratio of the related pages identified as abnormal pages.

[0144] As a possible implementation method of an embodiment of the present disclosure, the adjustment module 930 is used to: reduce the push probability of each page under the abnormal domain name in the page set when the page proportion of abnormal pages identified in the relevant pages is greater than the first proportion threshold; and / or, restore the push probability of each page under the abnormal domain name in the page set to the set probability when the page proportion of abnormal pages identified in the relevant pages is less than or equal to the first proportion threshold.

[0145] As a possible implementation method of the embodiment of the present disclosure, the adjustment module 930 is also used to: when it is identified that the proportion of abnormal pages in the relevant pages is greater than a first proportion threshold, determine each page under the abnormal domain name from the page set; identify abnormal pages for each page under the abnormal domain name in the page set; reduce the push probability of the abnormal page under the abnormal domain name to a first probability value, and reduce the push probability of the non-abnormal page under the abnormal domain name to a second probability value, wherein the second probability value is greater than the first probability value.

[0146] As a possible implementation method of the embodiment of the present disclosure, the adjustment module 930 is further used to: when it is identified that the proportion of abnormal pages in the relevant pages is greater than the first proportion threshold, determine the pages under the abnormal domain name from the page set; query the publisher of each page under the abnormal domain name in the page set to determine the pages belonging to the abnormal publisher; wherein the proportion of abnormal pages in the pages published by the abnormal publisher is greater than the second proportion threshold; for the pages belonging to the abnormal publisher, reduce the push probability to a third probability value, and for the pages that do not belong to the abnormal publisher, reduce the push probability to a fourth probability value, wherein the fourth probability value is greater than the third probability value.

[0147] As a possible implementation of the embodiment of the present disclosure, the page pushing device 900 further includes: a selection module and a pushing module.

[0148] Among them, the selection module is used to respond to the trigger word sent by the client and select candidate pages matching the trigger word from the page set; the push module is used to sort the candidate pages according to the push probability to select the pages with a set number of pages ranked first and push them to the client.

[0149] As a possible implementation method of the embodiment of the present disclosure, the monitoring module 920 is used to: determine the page ratio of abnormal pages in the first page when it is identified that there are abnormal pages in the first page; use the domain name of the first page as an abnormal domain name when the page ratio of abnormal pages in the first page is greater than a third ratio threshold; monitor the second page subsequently added to the page set based on the abnormal domain name to obtain the page with the abnormal domain name in the second page; and sample and select relevant pages from the pages with the abnormal domain name in the second page.

[0150] As a possible implementation of the embodiment of the present disclosure, the identification module 910 is used to: extract features from the set content of the first page to be identified or the related pages; wherein the set content includes one or more combinations of page text content, industry information, associated words, attributes of the publisher, and visitor access behavior information; and classify the extracted features using a classification model to determine whether it is an abnormal page based on the output of the classification model.

[0151] As a possible implementation method of an embodiment of the present disclosure, the recognition module 910 is also used to: extract the page text content for the first page to be identified or the related page; segment the page text content to determine the word vector of each word; fuse the word vectors of each word in the same sentence to obtain a sentence vector; and use the sentence vector of each sentence in the text part as a feature of the page text content.

[0152] As a possible implementation of the embodiment of the present disclosure, the page pushing device 900 further includes: an acquisition module, a determination module, and a deletion module.

[0153] Among them, the acquisition module is used to obtain the page text content of each page in the page set; the determination module is used to determine that the corresponding page is a dead link page when the page text content matches the set dead link keyword; the deletion module is used to delete the dead link page from the page set.

[0154] In summary, by monitoring the case where a new first page is added to the page set to be pushed, the first page is identified as an abnormal page; when it is identified that the first page is abnormal, the domain name of the first page is used as the abnormal domain name, and the second page subsequently added to the page set is monitored to select related pages with abnormal domain names from the second pages; the related pages are identified as abnormal pages; according to the page ratio of the identified abnormal pages in the related pages, the push probability of each page under the abnormal domain name in the page set is adjusted. Therefore, by identifying the abnormal pages under the abnormal domain name and adjusting the push probability of each page under the abnormal domain name in the page set according to the page ratio of the abnormal pages under the abnormal domain name, it is possible to manage and control the pages in the page set to be pushed that have low page quality and affect the user experience, thereby improving the user experience.

[0155] It should be noted that in the technical solutions disclosed herein, the collection, storage, use, processing, transmission, provision and disclosure of user personal information are all carried out with the user's consent, and are in compliance with relevant laws and regulations and do not violate public order and good morals.

[0156] In order to implement the above embodiments, the present disclosure also proposes an electronic device, comprising: at least one processor; and 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, and the instructions are executed by the at least one processor so that the at least one processor can execute the page push method of the above embodiments.

[0157] In order to implement the above embodiments, the present disclosure further proposes a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute the page pushing method described in the above embodiments.

[0158] In order to implement the above embodiments, the present disclosure further proposes a computer program product, including a computer program, which implements the page pushing method described in the above embodiments when executed by a processor.

[0159] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0160] Figure 10 A schematic block diagram of an example electronic device 1000 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 can 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 examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0161] like Figure 10 As shown, the device 1000 includes a computing unit 1001, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1002 or a computer program loaded from a storage unit 1008 into a random access memory (RAM) 1003. Various programs and data required for the operation of the device 1000 can also be stored in the RAM 1003. The computing unit 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0162] Various components in device 1000 are connected to I / O interface 1005, including an input unit 1006, such as a keyboard, mouse, etc.; an output unit 1007, such as various types of displays, speakers, etc.; a storage unit 1008, such as a magnetic disk, optical disk, etc.; and a communication unit 1009, such as a network card, modem, wireless communication transceiver, etc. The communication unit 1009 allows device 1000 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0163] The computing unit 1001 can be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1001 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1001 performs the various methods and processes described above, such as the page push method. For example, in some embodiments, the page push method can be implemented as a computer software program that is tangibly contained in a machine-readable medium, such as the storage unit 1008. In some embodiments, part or all of the computer program can be loaded and / or installed on the device 1000 via the ROM 1002 and / or the communication unit 1009. When the computer program is loaded into the RAM 1003 and executed by the computing unit 1001, one or more steps of the image processing method described above can be performed. Alternatively, in other embodiments, the computing unit 1001 can be configured to perform the page push method by any other appropriate means (e.g., by means of firmware).

[0164] Various embodiments of the systems and techniques described 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), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0165] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0166] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0167] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the 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 acoustic input, voice input, or tactile input).

[0168] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), the Internet, and a blockchain network.

[0169] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may also be a server in a distributed system or a server integrated with a blockchain.

[0170] It's important to note that artificial intelligence (AI) is the study of how computers can simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, and planning). This encompasses both hardware and software technologies. AI hardware technologies generally include sensors, specialized AI chips, cloud computing, distributed storage, and big data processing. AI software technologies primarily encompass computer vision, speech recognition, natural language processing, machine learning / deep learning, big data processing, and knowledge graphs.

[0171] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not a limitation herein.

[0172] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A page push method, comprising: When a new first page is detected in the set of pages to be pushed, abnormal page identification is performed on the first page; When an abnormality is identified in the first page, the domain name of the first page is used as an abnormal domain name, and a second page subsequently added to the page set is monitored, and the domain name of the second page is compared with the abnormal domain name to select a related page with the abnormal domain name from the second page; Identifying abnormal pages on the relevant pages; Adjusting the push probability of each page under the abnormal domain name in the page set according to the page ratio of the abnormal pages identified in the related pages includes: When the proportion of abnormal pages identified in the related pages is greater than a first ratio threshold, determining each page under the abnormal domain name from the page set; Identifying abnormal pages for each page under the abnormal domain name in the page set; Lowering the push probability of the abnormal page under the abnormal domain name to a first probability value, and lowering the push probability of the non-abnormal page under the abnormal domain name to a second probability value, wherein the second probability value is greater than the first probability value; or, When the proportion of abnormal pages identified in the related pages is greater than a first ratio threshold, determining each page under the abnormal domain name from the page set; querying the publisher of each page under the abnormal domain name in the page set to determine the pages belonging to the abnormal publisher; wherein the proportion of abnormal pages in the pages published by the abnormal publisher is greater than a second proportion threshold; For pages belonging to abnormal publishers, the push probability is reduced to a third probability value, and for pages not belonging to abnormal publishers, the push probability is reduced to a fourth probability value, wherein the fourth probability value is greater than the third probability value.

2. The method according to claim 1, wherein The adjusting the push probability of each page under the abnormal domain name in the page set according to the page proportion of the abnormal pages identified in the related pages further includes: When the proportion of abnormal pages identified in the related pages is less than or equal to the first ratio threshold, the push probability of each page under the abnormal domain name in the page set is restored to the set probability.

3. The method according to any one of claims 1 to 2, wherein The method further comprises: In response to a trigger word sent by a client, selecting a candidate page matching the trigger word from the page set; The candidate pages are sorted according to their push probabilities, so as to select a set number of pages ranked first and push them to the client.

4. The method according to any one of claims 1 to 2, wherein When the first page is identified as abnormal, the domain name of the first page is used as an abnormal domain name, and a second page subsequently added to the page set is monitored to select a related page having the abnormal domain name from the second page, including: When it is identified that there are abnormal pages in the first pages, determining a page ratio of the abnormal pages in the first pages; When the proportion of abnormal pages in the first pages is greater than a third ratio threshold, treating the domain name of the first page as an abnormal domain name; Based on the abnormal domain name, monitoring a second page subsequently added to the page set to obtain a page having the abnormal domain name in the second page; The relevant pages are sampled and selected from the pages having the abnormal domain name in the second pages.

5. The method according to any one of claims 1 to 2, wherein The performing abnormal page identification on the first page, or the performing abnormal page identification on the related pages, includes: Extracting features from the set content of the first page or the related page to be identified; wherein the set content includes one or more combinations of page text content, industry information, associated words, attributes of the publisher, and visitor behavior information; The extracted features are classified using a classification model to determine whether the page is an abnormal page based on an output of the classification model.

6. The method according to claim 5, wherein: The feature extraction for the setting content of the first page to be identified or the related page includes: Extracting page text content for the first page to be identified or the related page; Segment the page text content to determine the word vector for each word; Merge the word vectors of each word in the same sentence to obtain the sentence vector; The sentence vector of each sentence in the text part is used as the feature of the text content of the page.

7. The method according to any one of claims 1 to 2, wherein: The method further comprises: Obtaining the page text content of each page in the page set; If the page text content matches the set dead link keyword, determining that the corresponding page is a dead link page; Delete the dead link page from the page set.

8. A page pushing device, comprising: an identification module, configured to identify an abnormal page for a first page when detecting that the first page is newly added to the set of pages to be pushed; a monitoring module configured to, upon identifying an anomaly in the first page, use the domain name of the first page as an abnormal domain name, monitor a second page subsequently added to the page set, compare the domain name of the second page with the abnormal domain name, and select related pages having the abnormal domain name from the second page; The identification module is further used to identify abnormal pages of the relevant pages; An adjustment module, configured to adjust the push probability of each page under the abnormal domain name in the page set according to the proportion of abnormal pages identified in the related pages; The adjustment module is used to: When the proportion of abnormal pages identified in the related pages is greater than a first ratio threshold, determining each page under the abnormal domain name from the page set; Identifying abnormal pages for each page under the abnormal domain name in the page set; Lowering the push probability of the abnormal page under the abnormal domain name to a first probability value, and lowering the push probability of the non-abnormal page under the abnormal domain name to a second probability value, wherein the second probability value is greater than the first probability value; or, When the proportion of abnormal pages identified in the related pages is greater than a first ratio threshold, determining each page under the abnormal domain name from the page set; querying the publisher of each page under the abnormal domain name in the page set to determine the pages belonging to the abnormal publisher; wherein the proportion of abnormal pages in the pages published by the abnormal publisher is greater than a second proportion threshold; For pages belonging to abnormal publishers, the push probability is reduced to a third probability value, and for pages not belonging to abnormal publishers, the push probability is reduced to a fourth probability value, wherein the fourth probability value is greater than the third probability value.

9. The device according to claim 8, wherein The adjustment module is further used to: When the proportion of abnormal pages identified in the related pages is less than or equal to the first ratio threshold, the push probability of each page under the abnormal domain name in the page set is restored to the set probability.

10. The device according to any one of claims 8 to 9, wherein: The device further comprises: a selection module, configured to select, in response to a trigger word sent by a client, a candidate page matching the trigger word from the page set; The push module is used to sort the candidate pages according to their push probabilities, so as to select a set number of pages ranked first and push them to the client.

11. The device according to any one of claims 8 to 9, wherein: The monitoring module is used to: When it is identified that there are abnormal pages in the first pages, determining a page ratio of the abnormal pages in the first pages; When the proportion of abnormal pages in the first pages is greater than a third ratio threshold, treating the domain name of the first page as an abnormal domain name; Based on the abnormal domain name, monitoring a second page subsequently added to the page set to obtain a page having the abnormal domain name in the second page; The relevant pages are sampled and selected from the pages having the abnormal domain name in the second pages.

12. The device according to any one of claims 8 to 9, wherein: The identification module is used to: Extracting features from the set content of the first page or the related page to be identified; wherein the set content includes one or more combinations of page text content, industry information, associated words, attributes of the publisher, and visitor behavior information; The extracted features are classified using a classification model to determine whether the page is an abnormal page based on an output of the classification model.

13. The device according to claim 12, wherein The identification module is further used to: Extracting page text content for the first page to be identified or the related page; Segment the page text content to determine the word vector for each word; Merge the word vectors of each word in the same sentence to obtain the sentence vector; The sentence vector of each sentence in the text part is used as the feature of the text content of the page.

14. The device according to any one of claims 8 to 9, wherein: The device further comprises: An acquisition module, configured to acquire the page text content of each page in the page set; A determination module, configured to determine that a corresponding page is a dead link page when the page text content matches a set dead link keyword; The deletion module is used to delete the dead link page from the page set.

15. An electronic device, characterized in that: include: 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, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.

16. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 7.

17. A computer program product comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the method according to any one of claims 1 to 7.

Images