Display method, device and storage medium

Abstract

The embodiments of the present application provide a display method, device, and storage medium. In the embodiments of the present application, list-level features can be extracted separately from multiple separation positions included in the candidate content list; and the folding position in the candidate content list can be determined based on the list-level features to display the search results to the user. Among them, using list-level features as the evaluation basis can avoid the problem of poor fitting effect caused by factors such as insufficient features of individual candidate content or the same name of candidate content in traditional rule fitting schemes; determining the evaluation index for each separation position based on list-level features can avoid the problem of traditional rule fitting schemes requiring the writing of massive rules with complex logic, thereby improving evaluation efficiency. Therefore, the display scheme provided by the embodiments of the present application can display search results to users more accurately and concisely, thereby improving user experience and user decision-making efficiency for search results.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a display method, device, and storage medium. Background Art

[0002] Strong presentation is an important product feature in map search. Strong presentation refers to displaying the most relevant results to users in map search, while folding away results that are somewhat different from the user's search target to create a more concise display list.

[0003] At present, a discriminant model is usually used to determine the strong display position. Specifically, rule fitting can be performed based on the two POI features before and after the separation position to determine whether the separation position meets the strong display requirements. If a separation position that meets the requirements appears in the process of sequentially judging each separation position, the judgment operation of the subsequent separation position is stopped, and the current separation position is determined as the strong display position.

[0004] It can be seen that the strong display position determined by this solution is not accurate enough. Moreover, the judgment process usually involves many rules and complex logic, which leads to poor processing efficiency. Summary of the Invention

[0005] Various aspects of the present application provide a display method, device, and storage medium for displaying search results to users more accurately and concisely.

[0006] An embodiment of the present application provides a display method, including:

[0007] In response to the search instruction, determining a candidate content list;

[0008] extracting list-level features at each of the plurality of separation positions included in the candidate content list;

[0009] Determining the evaluation index corresponding to each of the plurality of separation positions based on a mapping relationship between list-level features and evaluation indexes;

[0010] selecting a folding position from the plurality of separation positions according to the evaluation index;

[0011] The candidate content list is folded according to the folding position to display the search results.

[0012] The present application also provides a display method, including:

[0013] In response to a search operation in the search interface, a search instruction is sent to the server;

[0014] receiving a candidate content list and corresponding folding positions provided by a server in response to the search instruction, wherein the folding positions are selected by the server based on evaluation indices corresponding to respective separation positions included in the candidate content list, wherein the evaluation indices are determined based on list-level features extracted from respective separation positions;

[0015] The candidate content list is folded according to the folding position and displayed in the candidate content display area.

[0016] An embodiment of the present application further provides a computing device, including a memory and a processor;

[0017] The memory is used to store one or more computer instructions;

[0018] The processor is coupled to the memory and configured to execute the one or more computer instructions for:

[0019] In response to the search instruction, determining a candidate content list;

[0020] extracting list-level features at each of the plurality of separation positions included in the candidate content list;

[0021] Determining the evaluation index corresponding to each of the plurality of separation positions based on a mapping relationship between list-level features and evaluation indexes;

[0022] selecting a folding position from the plurality of separation positions according to the evaluation index;

[0023] The candidate content list is folded according to the folding position to obtain a display result.

[0024] The embodiment of the present application further provides a terminal device, including a memory and a processor;

[0025] The memory is used to store one or more computer instructions;

[0026] The processor is coupled to the memory and configured to execute the one or more computer instructions for:

[0027] In response to a search operation in the search interface, a search instruction is sent to the server;

[0028] receiving a candidate content list and corresponding folding positions provided by a server in response to the search instruction, wherein the folding positions are selected by the server based on evaluation indices corresponding to respective separation positions included in the candidate content list, wherein the evaluation indices are determined based on list-level features extracted from respective separation positions;

[0029] The candidate content list is folded according to the folding position and displayed in the candidate content display area.

[0030] An embodiment of the present application further provides a computer-readable storage medium storing computer instructions. When the computer instructions are executed by one or more processors, the one or more processors are caused to execute the aforementioned display method.

[0031] In an embodiment of the present application, list-level features can be extracted from multiple separation positions in the candidate content list; and based on the list-level features, the folding position in the candidate content list can be determined to display the search results to the user. Accordingly, the display solution provided in an embodiment of the present application can produce at least one of the following technical effects:

[0032] 1. Using list-level features as the evaluation basis can avoid the problem of poor fitting results caused by factors such as insufficient features of individual candidate content or duplicate candidate content in traditional rule fitting solutions;

[0033] 2. Evaluate all the separator positions in the candidate content list to ensure that all separator positions in the candidate content list are evaluated, thereby more accurately determining the folding position;

[0034] 3. Determining the evaluation index for each separator position based on list-level features can avoid the problem of traditional rule-fitting solutions requiring the writing of massive rules with complex logic, thereby improving evaluation efficiency.

[0035] Therefore, the display solution provided in the embodiment of the present application can display search results to users more accurately and concisely, thereby improving user experience and user decision-making efficiency on search results. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0037] Figure 1 A flowchart of a display method provided by an exemplary embodiment of the present application;

[0038] Figure 2 A logical diagram of a display solution provided for an exemplary embodiment of the present application;

[0039] Figure 3 A flowchart of another display method provided by an exemplary embodiment of the present application;

[0040] Figure 4 A schematic diagram of a folding effect provided by an exemplary embodiment of the present application;

[0041] Figure 5 A schematic structural diagram of a computing device provided for another exemplary embodiment of the present application;

[0042] Figure 6 A schematic structural diagram of a terminal device provided as another exemplary embodiment of the present application. DETAILED DESCRIPTION

[0043] To make the purpose, technical solutions, and advantages of this application more clear, the technical solutions of this application will be clearly and completely described below in conjunction with the specific embodiments of this application and the corresponding drawings. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0044] At present, in the process of map search, the accuracy of the strong display position determined is insufficient and the processing efficiency is also poor. To this end, in some embodiments of the present application: list-level features can be extracted separately for multiple separation positions included in the candidate content list; and the folding position in the candidate content list can be determined based on the list-level features to display the search results to the user. Among them, using list-level features as the evaluation basis can avoid the problem of poor fitting effect caused by factors such as insufficient features of individual candidate content or the same name of candidate content in the traditional rule fitting scheme; evaluating all multiple separation positions in the candidate content list can ensure that all separation positions in the candidate content list get an evaluation opportunity, so that the folding position can be determined more accurately; determining the evaluation index for each separation position based on list-level features can avoid the problem of traditional rule fitting schemes requiring the writing of massive rules with complex logic, thereby improving evaluation efficiency. Therefore, the display scheme provided by the embodiment of the present application can display search results to users more accurately and concisely, thereby improving user experience and user decision-making efficiency for search results.

[0045] The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0046] Figure 1 This is a flow chart of a display method provided by an exemplary embodiment of the present application. Figure 2 This is a logical diagram of a display scheme provided by an exemplary embodiment of the present application. The display method can be executed by a data processing device, which can be implemented as a combination of software and / or hardware, and the data processing device can be integrated into a computing device. Figure 1 , the method comprising:

[0047] Step 100: In response to a search instruction, extract list-level features at multiple separation positions included in the candidate content list corresponding to the target client;

[0048] Step 101: Evaluate multiple separation positions based on list-level features to determine evaluation indices corresponding to the multiple separation positions.

[0049] Step 102: Select a folding position from a plurality of separation positions according to the evaluation index and notify the target client so that the target client can fold the candidate content list according to the folding position.

[0050] The display method provided in this embodiment can be applied to various search scenarios, such as map searches, searches on portal websites, and the like. This embodiment does not limit the application scenarios. In different application scenarios, the types of candidate content found may vary. For example, in a map search, the candidate content found may be points of interest (POIs), while in a portal website search scenario, the candidate content found may be URLs, applications, and the like.

[0051] The display method provided in this embodiment can be implemented by the server. In this case, in step 100, a search instruction initiated by the target client can be received, wherein the search instruction can include keywords. Based on this, candidate content that matches the keywords can be searched to obtain a candidate content list. The candidate content list can also be returned to the target client. It is worth noting that in this process, the searched candidate content can be sorted according to the current or future possible sorting methods to construct a candidate content list. For example, the click-through rate, quality score and correlation between each candidate content can be calculated separately, so as to determine the ranking score of each candidate content by weighted summation and other methods to determine the ranking between the candidate contents. The quality score can be evaluated from the perspectives of the content richness of the candidate content, the number of clicks, etc.

[0052] After the candidate content list is determined, in the subsequent processing, the order in the list will be kept unchanged, and the folding position will be determined in the candidate content list.

[0053] For this purpose, refer to Figure 1 and Figure 2In step 100, list-level features may be extracted for each of the multiple separator positions included in the candidate content list. Separator positions refer to the positions between adjacent candidate content items in the candidate content list. If the candidate content list contains N candidate content items, then the candidate content list contains N-1 separator positions. List-level features may be extracted from the global dimension of the candidate content list or the segment dimension of the candidate content list. Taking a map search scenario as an example, list-level features may include, but are not limited to, average relevance index, click coverage ratio, average click-through rate, click-through rate change rate at a separator position, or relevance change rate at a separator position. The average relevance index may be the average correlation between the candidate content in the list segment corresponding to the separator position and the search keyword. The click coverage ratio may be the ratio of the number of clicks on the candidate content in the list segment preceding the separator position to the total number of clicks on the entire candidate content list. The average click-through rate may be the average click-through rate of the candidate content in the list segment corresponding to the separator position. Of course, these are merely exemplary. List-level features may also utilize other features extracted from the global or segment dimensions of the list. This embodiment is not limited to these. For example, in a search scenario on a portal website, list-level features may also include the average activity level of a URL. In this embodiment, the list-level features extracted at different separation positions will not be exactly the same. Based on the list-level features, the list state of the candidate content list after being separated at a single separation position can be better characterized, thereby providing a more macro basis for accurately evaluating the separation effect at the separation position, without being limited by the insufficient features or feature defects of a single candidate content.

[0054] In step 101, the multiple separation positions in the candidate content list can be evaluated based on the list-level features extracted at the multiple separation positions to determine the evaluation index corresponding to each of the multiple separation positions. In an optional implementation scheme, the evaluation index corresponding to each of the multiple separation positions can be determined based on the mapping relationship between the list-level features and the evaluation index. In other words, the evaluation index that is compatible with the list-level features at the multiple separation positions is determined. The mapping relationship between the list-level features and the evaluation index can be constructed based on expert experience or through summarizing and learning sample data. Of course, other implementation methods can also be used to perform the evaluation operation in this embodiment, for example, constructing a linear or nonlinear equation and substituting the list-level features into the equation to generate the evaluation index, but this embodiment is not limited to this. Therefore, there is no need to write different, diverse, and complex rule logic for different separation positions as in traditional rule fitting schemes.

[0055] Additionally, the evaluation index can be used to assess the display effect of folding the candidate content list according to the current separator position. Furthermore, in this embodiment, all multiple separator positions included in the candidate content list can be evaluated. This allows each separator position to be evaluated, thus avoiding missing out on separator positions with excellent display effects.

[0056] Based on this, in step 102, a folding position can be selected from the multiple partition positions based on their corresponding evaluation indices. Optionally, the partition position with the highest evaluation index can be selected as the folding position from the multiple partition positions. Of course, this embodiment is not limited to this. For example, the rearmost partition position among the M partition positions with the highest evaluation indices can be selected as the folding position, etc. In this way, the folding position prediction problem can be transformed into a ranking problem, allowing for more efficient and concise determination of the folding position. After the folding position is determined, the folding position can be notified to the target client, allowing the target client to fold the candidate content list according to the folding position.

[0057] Accordingly, the display solution provided in this embodiment can produce at least one of the following technical effects:

[0058] 1. Using list-level features as the evaluation basis can avoid the problem of poor fitting results caused by factors such as insufficient features of individual candidate content or duplicate candidate content in traditional rule fitting solutions;

[0059] 2. Evaluate all the separator positions in the candidate content list to ensure that all separator positions in the candidate content list are evaluated, thereby more accurately determining the folding position;

[0060] 3. Determining the evaluation index for each separator position based on list-level features can avoid the problem of traditional rule-fitting solutions requiring the writing of massive rules with complex logic, thereby improving evaluation efficiency.

[0061] Therefore, the display solution provided in the embodiment of the present application can display search results to users more accurately and concisely, thereby improving user experience and user decision-making efficiency on search results.

[0062] In the above or following embodiments, taking the kth separator position contained in the candidate content list as an example, the candidate content list can be split into two candidate content groups according to the kth separator position; list-level features are extracted from the two candidate content groups respectively as the list-level features corresponding to the kth separator position.

[0063] If the candidate content list contains N candidate contents, then k=1, 2, ..., N-1, and the two candidate content groups will be (P1, P2, ..., P k ), (Pk+1 、P k+2 ,…,P N ), P k For example, if N=10 and k=4, the candidate content group before the k-th separation position can be shown in Table 1 below:

[0064] Geographic location 1 Geographic location 2 Geographic location 3 Geographic location 4

[0065] The candidate content group located after the kth separation position may be shown in Table 2 below:

[0066] Geographic location 5 Geographic location 6 Geographic location 7 Geographic location 8 Geographic location 9 Geographic location 10

[0067] It can be seen that the two candidate content groups are actually fragments of the candidate content list, that is, by separating the candidate content list according to the k-th separation position, two candidate content list fragments can be obtained. In this embodiment, list-level features can be extracted separately under the two candidate content list fragments. Here, the list-level features extracted under a single candidate content group are consistent with the list-level features mentioned in the aforementioned embodiment. For example, in a map search scenario, features such as the average relevance index, click coverage ratio, average click rate, click rate change rate at the separation position, or relevance change rate at the separation position can be extracted from a single candidate content group. The list-level features under another candidate content group can be combined to generate the list-level features of the k-th separation position under the candidate content list.

[0068] One implementation scheme for extracting list-level features under a single candidate content group may be: obtaining the attribute features corresponding to each candidate content contained in the candidate content group; and calculating the list-level features corresponding to the candidate content group based on the attribute features corresponding to each candidate content. The attribute features are adapted to the list-level features, and the list-level features may be obtained by further processing the attribute features. Taking the map search scenario as an example, the attribute features may include but are not limited to relevance, click-through rate, number of clicks, etc., but this embodiment is not limited to this. In actual applications, the search log can be read, and the attribute features of each candidate content can be counted based on the search log, wherein the search log records the user behavior data during the historical search process, and these user behavior data can directly or indirectly reflect the attribute features of the candidate content.

[0069] Following Tables 1 and 2 above, the attribute features corresponding to each candidate content can be expressed as Tables 3 and 4 below:

[0070] Table 3

[0071] Geographic location 1 Correlation 0.8 Click-through rate 0.8 Number of clicks: 15 Geographic location 2 Correlation 0.6 Click-through rate 0.9 Number of clicks: 20 Geographic location 3 Correlation 0.7 Click-through rate 0.6 Number of clicks: 14 Geographic location 4 Correlation 0.5 Click-through rate 0.7 Number of clicks: 10

[0072] Table 4

[0073] Geographic location 5 Correlation 0.5 Click-through rate 0.5 Number of clicks: 10 Geographic location 6 Correlation 0.6 Click-through rate 0.3 Number of clicks: 10 Geographic location 7 Correlation 0.5 Click-through rate 0.6 Number of clicks: 9 Geographic location 8 Correlation 0.3 Click-through rate 0.4 Number of clicks: 11 Geographic location 9 Correlation 0.2 Click-through rate 0.2 Number of clicks: 3 Address location 10 Correlation 0.3 Click rate 0.1 Number of clicks: 1

[0074] For example, for the candidate content groups (P1, P2, ..., P k ), taking the average click rate feature in the aforementioned list-level feature as an example: the candidate content groups (P1, P2, ..., P k ), and then calculate the click-through rate of the k candidate contents in the candidate content group, and use it as the average click-through rate feature. Taking the click-through rate change rate feature at the separation position in the aforementioned list-level feature as an example: the candidate content group (P1, P2, ..., P k ) k Click-through rate and candidate content groups (P k+1 、P k+2 ,…,P N ) k+1 The click rate of the kth position is calculated as the click rate change rate feature at the kth separation position. According to the same implementation scheme, the candidate content group (P k+1 、P k+2 ,…,P N ) corresponding to various list-level features.

[0075] Following the above table, the list-level features of the k-th separator position can be expressed as Table 5:

[0076]

[0077] Based on this, in this embodiment, the candidate content list can be hypothetically divided into two candidate content groups based on the separation position, and list-level features of the two candidate content groups can be extracted. The list-level features extracted under this dimension can characterize the contextual differences and other states between the two candidate content groups. This allows for a more accurate, reasonable, and comprehensive evaluation of the display effects of each separation position when using these list-level features as the evaluation basis, thereby effectively improving the accuracy and rationality of the folding position. Of course, in this embodiment, other implementation methods can also be used to extract list-level features, and this embodiment is not limited to this.

[0078] In the above or following embodiments, when determining evaluation indices for multiple separator positions, list-level features for the multiple separator positions may be input into an evaluation model. Based on the list-level features for the multiple separator positions and the mapping relationship between the list-level features and the evaluation indices, the evaluation model outputs evaluation indices corresponding to each of the multiple separator positions. The evaluation model may employ a machine learning model, such as a support vector machine (SVM), a gradient descent tree (GBDT), a logistic regression (LR), or other learning-to-rank (LTR) model.

[0079] Based on the evaluation model, in this embodiment, the problem of determining the folding position can be transformed into a problem of sorting multiple separation positions based on display effects, thereby more accurately estimating the folding position in the candidate content list.

[0080] The following details the training process of the evaluation model:

[0081] Get display samples;

[0082] Extract list-level features separately under multiple separation positions where the sample is displayed;

[0083] Annotating a first evaluation index for a list-level feature corresponding to a folding position among a plurality of separation positions included in the display sample;

[0084] Label the second evaluation index for the list-level features corresponding to other separation positions included in the display sample;

[0085] The list-level features and evaluation indices corresponding to the multiple separation positions contained in the display sample are input into the evaluation model so that the evaluation model can learn the mapping relationship between the list-level features and the evaluation indices.

[0086] The above training process is described from the perspective of a single display sample. It should be understood that in practical applications, a large number of display samples can be input into the evaluation model, allowing the evaluation model to continuously optimize the mapping relationship between the learned list-level features and the evaluation index. Furthermore, the process of extracting list-level features from multiple separation positions within a display sample in this embodiment can be referenced to the description in the previous embodiment and will not be repeated here.

[0087] During the training process, the folding positions in the display samples can be manually determined. Alternatively, during the sample selection phase, display samples that meet the preset display effect requirements can be selected. In this case, the folding positions provided in the display samples can be directly used. In this embodiment, the method for determining the folding positions in the display samples is not limited.

[0088] In this embodiment, taking the display sample containing N candidate contents as an example, the sample format of the display sample at the kth separation position can be (label label, P1~P k , P k+1 ~P N ), where label corresponds to the first evaluation index or the second evaluation index. Optionally, the first evaluation index can be 1 and the second evaluation index can be 0. Thus, continuing with the above example, if the kth separator position is a folded position, the sample format corresponding to the kth separator position can be (1, P1~P k , P k+1 ~P N), where P1~P k , P k+1 ~P N They represent the list-level features corresponding to the two candidate content groups obtained after dividing the display sample according to the kth separation position. In the sample forms corresponding to other separation positions of the display sample, in addition to the change of the candidate content grouping, the value of label will also become 0.

[0089] In this way, by inputting a massive amount of display samples into the evaluation model, the mapping relationship between list-level features and evaluation indices in the evaluation model can be continuously optimized. From the perspective of the evaluation model, the display scheme provided in the aforementioned embodiment can output evaluation indices for each of the multiple separation positions in the candidate content list. These evaluation indices will fall within the interval [0, 1]. Since the list-level features corresponding to different separation positions are not identical, the corresponding evaluation indices will also be different. Based on this, the evaluation indices can be used to represent the display effect of the candidate content list after folding at different separation positions, thereby accurately identifying the folding position.

[0090] Based on this, this embodiment uses machine learning to transform the problem of determining the fold position into the problem of sorting multiple separation positions based on display effects, thereby more accurately estimating the fold position within the candidate content list. Compared to traditional rule-fitting solutions, this eliminates the labor and material resources required to write numerous complex rules, effectively improving the accuracy and efficiency of processing results.

[0091] Figure 3 A flow chart of another display method provided by an exemplary embodiment of the present application. The display method can be executed by a data processing device, which can be implemented as a combination of software and / or hardware, and the data processing device can be integrated into a terminal device. Figure 3 , the method may include:

[0092] Step 300: Display the search interface;

[0093] Step 301: In response to a search operation in the search interface, a search instruction is sent to the server.

[0094] Step 302: Receive a candidate content list and corresponding folding positions provided by the server in response to a search instruction, wherein the folding positions are selected by the server based on evaluation indices corresponding to respective separation positions in the candidate content list, wherein the evaluation indices are determined based on list-level features extracted from respective separation positions;

[0095] Step 303: After folding the candidate content list according to the folding position, the list is displayed in the candidate content display area.

[0096] Figure 4 This is a schematic diagram of a folding effect provided by an exemplary embodiment of the present application. Figure 4 , the terminal device can provide a user interaction interface, which includes a search interface. In this embodiment, the candidate content display area can also be displayed in the user interaction interface, and of course other display areas can also be displayed, such as Figure 4 The map display area shown is adapted to the display requirements of the application scenario. These displayed contents can be presented in the search interface. Of course, they can also be presented in other interfaces other than the search interface in the user interaction interface. This is not limited in this embodiment. The user can perform a search operation in the search interface. To this end, a search control can be displayed in the search interface for the user to enter keywords to search. In this embodiment, keywords can be provided to the server so that the server can determine a list of candidate content based on the keywords. For specific technical details, please refer to the description in the previous embodiment and will not be repeated here.

[0097] In addition, it is worth noting that in this embodiment, based on the candidate content list having been obtained, the server may no longer determine the folding position. Instead, the terminal device in this embodiment may execute the relevant processing logic for determining the folding position. In this case, this embodiment may execute the following processing logic: extracting list-level features for each of the multiple separator positions included in the candidate content list; evaluating the multiple separator positions based on the list-level features to determine the evaluation index corresponding to each of the multiple separator positions; and selecting a folding position from the multiple separator positions based on the evaluation index. This eliminates the need to receive the folding position returned by the server.

[0098] For the technical details on determining the folding position, please refer to the description in the previous embodiment, which will not be repeated here.

[0099] In either case, in this embodiment, the candidate content can be folded according to the folding position, and the folded candidate content list can be displayed in the candidate content display area. Specifically, in this embodiment, the candidate content list can be split into two candidate content groups according to the folding position; the candidate content group located after the folding position is hidden, and only the candidate content group located before the folding position is displayed. Figure 4 If the kth separator position is used as the folding position, the candidate content after the kth separator position in the candidate content list can be folded, wherein the folding process can be to hide the candidate content after the folding position. In actual applications, an expansion service can also be provided in the display interface of the search results, refer to Figure 4 For example, an operable control of "click to query more content" may be provided to support the expansion service.

[0100] Based on this, in this embodiment, the terminal device side can fold the candidate content list more efficiently and accurately, thereby achieving a strong display of the candidate content, obtaining more concise search results, and improving the user's decision-making efficiency.

[0101] It should be noted that some of the processes described in the above embodiments and the accompanying drawings include multiple operations that appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The sequence numbers of the operations, such as 101, 102, etc., are merely used to distinguish between different operations and do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel.

[0102] Figure 5 This is a schematic diagram of a computing device provided by another exemplary embodiment of the present application. Figure 5 As shown, the computing device includes a memory 50 and a processor 51 .

[0103] The processor 51 is coupled to the memory 50 and is configured to execute the computer program in the memory 50 to:

[0104] In response to the search instruction, determining a candidate content list;

[0105] Extracting list-level features at multiple separation positions contained in the candidate content list;

[0106] Determine the evaluation index corresponding to each of the plurality of separation positions based on the mapping relationship between the list-level features and the evaluation index;

[0107] selecting a folding position from a plurality of separation positions according to the evaluation index;

[0108] The candidate content list is folded according to the folding position to display the search results.

[0109] In an optional embodiment, when the processor 51 extracts list-level features at multiple separation positions included in the candidate content list, it is configured to:

[0110] Splitting the candidate content list into two candidate content groups according to the first separation position;

[0111] Extracting list-level features from the two candidate content groups as list-level features corresponding to the first separator position;

[0112] The first separation position is any one of the multiple separation positions.

[0113] In an optional embodiment, when the processor 51 extracts list-level features from two candidate content groups respectively, it is configured to:

[0114] Obtaining attribute features corresponding to each candidate content included in the two candidate content groups;

[0115] According to the attribute features corresponding to each candidate content, list-level features corresponding to the two candidate content groups are calculated respectively.

[0116] In an optional embodiment, the attribute characteristics include one or more of relevance, click-through rate, and number of clicks; the list-level characteristics include one or more of average relevance index, click coverage ratio, average click-through rate, click-through rate change rate at the separation position, and relevance change rate at the separation position.

[0117] In an optional embodiment, when determining the evaluation index corresponding to each of the plurality of separation positions based on the mapping relationship between the list-level features and the evaluation index, the processor 51 is configured to:

[0118] Inputting list-level features under multiple separation positions into the evaluation model;

[0119] In the evaluation model, based on the list-level features at the multiple separation positions and the mapping relationship between the list-level features and the evaluation index, the evaluation index corresponding to each of the multiple separation positions is output.

[0120] In an optional embodiment, the processor 51 is configured to:

[0121] Get display samples;

[0122] Extract list-level features separately under multiple separation positions where the sample is displayed;

[0123] Annotating a list-level feature corresponding to a folding position among a plurality of separation positions included in the display sample as a first evaluation index;

[0124] The list-level features corresponding to other separation positions contained in the display sample are marked as the second evaluation index;

[0125] The list-level features and evaluation indices corresponding to the multiple separation positions contained in the display sample are input into the evaluation model so that the evaluation model can learn the mapping relationship between the list-level features and the evaluation indices.

[0126] In an optional embodiment, the first evaluation index is 1 and the second evaluation index is 0.

[0127] In an optional embodiment, when selecting a folding position from a plurality of separation positions according to the evaluation index, the processor 51 is configured to:

[0128] From the plurality of separation positions, the separation position having the highest evaluation index is selected as the folding position.

[0129] In an optional embodiment, the search instruction is a map search instruction, and the candidate content list is a point of interest (POI) list.

[0130] Further, if Figure 5 As shown, the computing device also includes: a communication component 52, a power supply component 53 and other components. Figure 5 Only some components are shown schematically, and it does not mean that the computing device only includes Figure 5 Components shown.

[0131] It is worth noting that the technical details in the above-mentioned embodiments of the computing device can be referred to the relevant description in the aforementioned method embodiment. In order to save space, they will not be repeated here, but this should not cause any loss of the scope of protection of this application.

[0132] Figure 6 This is a schematic diagram of the structure of a terminal device provided by another exemplary embodiment of the present application. Figure 6 , the terminal device may include a memory and a processor;

[0133] The memory is used to store one or more computer instructions;

[0134] The processor is coupled to the memory and is configured to execute one or more computer instructions for:

[0135] In response to a search operation in the search interface, a search instruction is sent to the server;

[0136] receiving a candidate content list and corresponding folding positions provided by a server in response to a search instruction, wherein the folding positions are selected by the server based on evaluation indices corresponding to respective plurality of separation positions included in the candidate content list, wherein the evaluation indices are determined based on list-level features extracted from respective plurality of separation positions;

[0137] After the candidate content list is folded according to the folding position, it is displayed in the candidate content display area.

[0138] In an optional embodiment, during the process of folding the candidate content list according to the folding position, the processor 61 may be configured to:

[0139] Split the candidate content list into two candidate content groups according to the folding position;

[0140] Hide the candidate content group located after the collapsed position.

[0141] Further, if Figure 6 As shown, the terminal device also includes: a communication component 62, a display 63, a power supply component 64, an audio and video component 55 and other components. Figure 6Only some components are shown schematically, which does not mean that the terminal equipment only includes Figure 6 Components shown.

[0142] It is worth noting that the technical details in the above-mentioned embodiments of the terminal device can be referred to the relevant description in the aforementioned method embodiment. In order to save space, they will not be repeated here, but this should not cause any loss of the protection scope of this application.

[0143] Accordingly, an embodiment of the present application also provides a computer-readable storage medium storing a computer program, which, when executed, can implement the steps in the above method embodiment that can be executed by a computing device or a terminal device.

[0144] above Figure 4 and 5 The memory in the computing platform is used to store computer programs and can be configured to store various other data to support operations on the computing platform. Examples of such data include instructions for any application or method operating on the computing platform, contact data, phone book data, messages, pictures, videos, etc. The memory can be implemented by any type of volatile or non-volatile storage device or a combination of them, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0145] above Figure 4 and 5 The communication component in is configured to facilitate wired or wireless communication between the device where the communication component is located and other devices. The device where the communication component is located can access a wireless network based on a communication standard, such as WiFi, 2G, 3G, 4G / LTE, 5G and other mobile communication networks, or a combination thereof. In an exemplary embodiment, the communication component receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0146] above Figure 5The display in the embodiment includes a screen, which may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation.

[0147] above Figure 4 and 5 The power supply component in a device provides power to various components of the device in which the power supply component is located. The power supply component may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device in which the power supply component is located.

[0148] above Figure 5 The audio component in the device may be configured to output and / or input audio signals. For example, the audio component includes a microphone (MIC), and when the device where the audio component is located is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode, the microphone is configured to receive an external audio signal. The received audio signal may be further stored in a memory or sent via a communication component. In some embodiments, the audio component also includes a speaker for outputting audio signals.

[0149] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0150] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0151] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0152] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0153] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0154] Memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0155] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0156] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0157] The foregoing is merely an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included in the protection scope of the present application.

Claims

1. A display method, characterized in that: include: In response to a search instruction initiated by a target client, extracting list-level features at each of a plurality of separation positions included in a candidate content list corresponding to the target client, wherein the separation positions are positions between adjacent candidate content in the candidate content list, and the list-level features are features extracted from a global dimension or a segment dimension of the candidate content list; Evaluate the plurality of separation positions based on the list-level features to determine evaluation indices corresponding to the plurality of separation positions; According to the evaluation index, a folding position is selected from the plurality of separation positions and notified to the target client, so that the target client folds the candidate content list according to the folding position.

2. The method according to claim 1, characterized in that The extracting list-level features at the multiple separation positions included in the candidate content list includes: Splitting the candidate content list into two candidate content groups according to the first separation position; extracting list-level features from the two candidate content groups respectively as list-level features corresponding to the first separation position; The first separation position is any one of the multiple separation positions.

3. The method according to claim 2, characterized in that The extracting list-level features from the two candidate content groups respectively includes: Obtaining attribute features corresponding to each candidate content included in the two candidate content groups; List-level features corresponding to the two candidate content groups are calculated respectively according to the attribute features corresponding to the respective candidate contents.

4. The method according to claim 3, characterized in that The attribute characteristics include one or more of relevance, click-through rate, and number of clicks; the list-level characteristics include one or more of average relevance index, click coverage ratio, average click-through rate, click-through rate change rate at the separation position, and relevance change rate at the separation position.

5. The method according to claim 1, characterized in that The evaluating the plurality of separation positions based on the list-level features to determine evaluation indices corresponding to the plurality of separation positions includes: inputting the list-level features under the plurality of separation positions into an evaluation model; In the evaluation model, based on the list-level features at the multiple separation positions and the mapping relationship between the list-level features and the evaluation index, the evaluation index corresponding to each of the multiple separation positions is output.

6. The method according to claim 5, characterized in that The training process of the evaluation model includes: Get display samples; Extracting list-level features respectively at multiple separation positions included in the display sample; Marking a first evaluation index for a list-level feature corresponding to a folding position among a plurality of separation positions included in the display sample; Marking second evaluation indexes for list-level features corresponding to other separation positions included in the display sample; The list-level features and evaluation indices corresponding to the plurality of separation positions included in the display sample are input into the evaluation model so that the evaluation model can learn the mapping relationship between the list-level features and the evaluation indices.

7. The method according to claim 6, characterized in that The first evaluation index is 1, and the second evaluation index is 0.

8. The method according to claim 1, characterized in that The step of selecting a folding position from the plurality of separation positions according to the evaluation index comprises: From the plurality of separation positions, a separation position having the highest evaluation index is selected as the folding position.

9. The method according to claim 1, characterized in that The search instruction is a map search instruction, and the candidate content list is a point of interest (POI) list.

10. A display method, characterized in that: include: In response to a search operation in the search interface, a search instruction is sent to the server; receiving a candidate content list and corresponding folding positions provided by a server in response to the search instruction, wherein the folding positions are selected by the server based on evaluation indices corresponding to respective separation positions included in the candidate content list, wherein the evaluation indices are determined based on list-level features extracted from the respective separation positions, wherein the separation positions are positions between adjacent candidate contents in the candidate content list, and the list-level features are features extracted from a global dimension or a segment dimension of the candidate content list; The candidate content list is folded according to the folding position and displayed in the candidate content display area.

11. The method according to claim 10, characterized in that The folding of the candidate content list according to the folding position includes: splitting the candidate content list into two candidate content groups according to the folding position; The candidate content group located after the folding position is hidden.

12. A computing device, characterized in that including memory and processor; The memory is used to store one or more computer instructions; The processor is coupled to the memory and configured to execute the one or more computer instructions for: In response to a search instruction initiated by a target client, extracting list-level features at each of a plurality of separation positions included in a candidate content list corresponding to the target client, wherein the separation positions are positions between adjacent candidate content in the candidate content list, and the list-level features are features extracted from a global dimension or a segment dimension of the candidate content list; Evaluate the plurality of separation positions based on the list-level features to determine evaluation indices corresponding to the plurality of separation positions; According to the evaluation index, a folding position is selected from the plurality of separation positions and notified to the target client, so that the target client folds the candidate content list according to the folding position.

13. A terminal device, characterized in that: including memory and processor; The memory is used to store one or more computer instructions; The processor is coupled to the memory and configured to execute the one or more computer instructions for: In response to a search operation in the search interface, a search instruction is sent to the server; receiving a candidate content list and corresponding folding positions provided by a server in response to the search instruction, wherein the folding positions are selected by the server based on evaluation indices corresponding to respective separation positions included in the candidate content list, wherein the evaluation indices are determined based on list-level features extracted from the respective separation positions, wherein the separation positions are positions between adjacent candidate contents in the candidate content list, and the list-level features are features extracted from a global dimension or a segment dimension of the candidate content list; The candidate content list is folded according to the folding position and displayed in the candidate content display area.

14. A computer-readable storage medium storing computer instructions, characterized in that: When the computer instructions are executed by one or more processors, the one or more processors are caused to execute the display method according to any one of claims 1 to 11.

Images