House resource sorting method, electronic equipment, storage medium and program product

Abstract

The invention provides a housing resource sorting method, electronic equipment, a readable storage medium and a computer program product. The housing resource sorting method comprises the steps of determining a plurality of target housing resources in response to a housing resource search request of a target user; user portrait features of the target user and house resource attribute features of the target house resource are obtained, the user portrait features comprise user preference features and user behavior features, the user preference features comprise entities preferred by the target user, and the user behavior features comprise entities of historical behavior effects of the target user; the house attribute features comprise entities representing house attributes; determining an entity in the user preference feature, an entity in the user behavior feature and an entity in the house resource attribute feature in the knowledge graph as target entities, and obtaining a relation chain between the target entities; generating house resource sorting features based on the relation chain; and determining recommendation values of the target housing resources based on the housing resource sorting features, and sorting the plurality of target housing resources based on the recommendation values.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and more particularly to a housing source sorting method, electronic equipment, storage medium and program product. Background Art

[0002] Some property search pages can sort the properties to be displayed before displaying them, and can generate different sorting results according to the preferences of different users to obtain the display priority of the properties to be displayed, so as to make personalized content recommendations based on the user's preferences.

[0003] Some current house ranking models can estimate the probability of a house being clicked by a user and sort the houses according to the probability of being clicked. However, current house ranking models often need to be trained with a large number of user behavior data samples, so that the house ranking models learn the implicit expression relationship between user behavior and houses. This kind of learning cannot be controlled and the learning effect is poor. Summary of the invention

[0004] The present disclosure provides a housing source sorting method, electronic device, storage medium and program product.

[0005] According to one aspect of the present disclosure, a method for sorting housing sources is provided, comprising: In response to a target user's housing search request, determining a plurality of target housing sources; Obtaining user portrait features of the target user and property attribute features of the target property, wherein the user portrait features include user preference features and user behavior features, wherein the user preference features include entities preferred by the target user, the user behavior features include entities affected by the historical behavior of the target user, and the property attribute features include entities representing the properties of the property; Determine in the knowledge graph the entities in the user preference feature, the entities in the user behavior feature, and the entities in the property attribute feature as target entities, and obtain the relationship chain between the target entities; Generate a property sorting feature based on the relationship chain; The recommendation value of the target house is determined based on the house ranking feature, and the multiple target houses are ranked based on the recommendation value.

[0006] According to at least one embodiment of the housing source ranking method of the present disclosure, the type of the housing source ranking feature includes at least one of a binary feature, an enumeration feature, a quantity feature, and a vector feature; generating the housing source ranking feature based on the relationship chain includes: Query whether there is a relationship chain between the two target entities to obtain an existence query result, and generate the binary feature based on the existence query result, wherein the binary feature is used to indicate whether there is a relationship chain between the two target entities; and / or, Obtaining the names of each association relationship included in the relationship chain between the two target entities, and generating the enumeration class feature based on the acquired names, wherein the enumeration class feature is used to represent the names of each association relationship that associates the two target entities; and / or, Counting the number of each association relationship included in the relationship chain between the two target entities, and generating the quantity feature based on the counted number, wherein the quantity feature is used to represent the number of association relationships that associate the two target entities; and / or, The vector data in each association relationship included in the relationship chain between the two target entities is queried, and the queried vector data is used as the vector class feature.

[0007] According to at least one embodiment of the present disclosure, the housing source sorting method further includes: In response to an entity represented in the user portrait not being found in the knowledge graph, the entity that cannot be found is stored in the knowledge graph as a newly added entity, and the association relationship between the newly added entity and other entities in the knowledge graph is recorded in the knowledge graph.

[0008] According to the housing source ranking method of at least one embodiment of the present disclosure, the type of association relationship includes at least one of a subordinate relationship, a distance relationship, a common category relationship and a common attribute relationship, and the association relationship between the newly added entity and other entities in the knowledge graph is recorded in the knowledge graph, including: querying in the knowledge graph whether the newly added entity is an attribute of another entity in the knowledge graph, and if the newly added entity is an attribute of another entity in the knowledge graph, generating the subordinate relationship corresponding to the newly added entity based on the newly added entity and the queried another entity in the knowledge graph; and / or, Obtaining the attribute of the newly added entity, and querying in the knowledge graph whether the attribute of the newly added entity is another entity in the knowledge graph, and when the attribute of the newly added entity is another entity in the knowledge graph, generating the subordinate relationship corresponding to the newly added entity based on the newly added entity and the queried another entity in the knowledge graph; and / or, Obtaining a distance between the newly added entity and another entity in the knowledge graph, and when the obtained distance is less than a distance threshold, generating the distance relationship corresponding to the newly added entity based on the obtained distance; and / or, Obtaining the category of the newly added entity, and searching for entities of the same category in the knowledge graph, and when entities of the same category are found, generating the common category relationship corresponding to the newly added entity based on the category of the newly added entity; and / or, Entities having the same attributes as the newly added entity are queried in the knowledge graph, and when the entities having the same attributes are found, the common attribute relationship corresponding to the newly added entity is generated based on the attributes of the newly added entity.

[0009] According to the housing source ranking method of at least one embodiment of the present disclosure, the entity in the user behavior feature includes housing sources, the type of the user behavior feature includes at least one of a selection number feature, a selection sequence feature, a search number feature, a search sequence feature, and a re-search feature, and the entity in the user preference feature includes a search condition; obtaining the user portrait feature of the target user includes: Acquire the historical record information of the target user, wherein the historical record information includes at least one of the target user's historical search records for housing resources and historical click records for housing resources; Extracting the properties selected by the target user in a target time period and the number of times they were selected from the records of the historically clicked properties, generating a selection number feature based on each selected property and the number of times it was selected, and generating a selection sequence feature based on the target time period and each selected property; and / or, Extracting the houses searched by the target user in the target time period and the number of times they were searched from the records of the historical house searches, generating a search number feature based on each searched house and the number of times they were searched, and generating a search sequence feature based on the target time period and each searched house; and / or, Extracting the time interval between two searches of the same house by the target user from the historical house search records, and generating the re-search feature based on the two searches of the same house and the time interval; and / or, The target user's preference score for the search condition is obtained based on the historical records of house search, and a user preference feature is generated based on the search condition and its preference score.

[0010] According to at least one embodiment of the present disclosure, a method for ranking housing sources, determining a recommendation value of the target housing source based on the housing source ranking feature, includes: Transmitting the user portrait features, the property attribute features of the plurality of target properties, and the property ranking features to a recommendation module; Obtain recommendation values ​​of the multiple target properties returned by the recommendation module.

[0011] According to at least one embodiment of the present disclosure, a method for sorting housing sources is provided, wherein a plurality of target housing sources are determined, including: Obtaining search conditions in the search request of the target user; The house in the house database is searched based on the search condition, and the searched house is determined as the target house.

[0012] According to another aspect of the present disclosure, an electronic device is provided, comprising: a memory storing execution instructions; and a processor executing the execution instructions stored in the memory, so that the processor executes the housing source sorting method of any embodiment of the present disclosure.

[0013] According to another aspect of the present disclosure, a readable storage medium is provided, in which execution instructions are stored. When the execution instructions are executed by a processor, they are used to implement the housing source sorting method of any embodiment of the present disclosure.

[0014] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the housing source sorting method according to any embodiment of the present disclosure is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings illustrate exemplary embodiments of the present disclosure and together with the description serve to explain the principles of the present disclosure. These drawings are included to provide a further understanding of the present disclosure and are incorporated in and constitute a part of this specification.

[0016] Figure 1 It is a flowchart of a method for sorting housing resources according to an embodiment of the present invention.

[0017] Figure 2 It is a schematic diagram of a knowledge graph of an embodiment of the present disclosure.

[0018] Figure 3 It is a flowchart of a method for sorting housing resources according to an embodiment of the present invention.

[0019] Figure 4 It is a flowchart of a method for sorting housing resources according to an embodiment of the present invention.

[0020] Figure 5 It is a flowchart of a method for sorting housing resources according to an embodiment of the present invention.

[0021] Figure 6 It is a flowchart of a method for sorting housing resources according to an embodiment of the present invention.

[0022] Figure 7 It is a flowchart of a method for sorting housing resources according to an embodiment of the present invention.

[0023] Figure 8 It is a flowchart of a method for sorting housing resources according to an embodiment of the present invention.

[0024] Fig. 9 It is a schematic block diagram of the structure of a sorting device according to an embodiment of the present disclosure.

[0025] Fig.10 It is a schematic block diagram of the structure of an electronic device equipped with a sorting device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0026] The present disclosure is further described in detail below in conjunction with the accompanying drawings and examples. It is understood that the specific examples described herein are only used to explain the relevant content, rather than to limit the present disclosure. It should also be noted that, for ease of description, only the parts related to the present disclosure are shown in the accompanying drawings.

[0027] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in the present disclosure can be combined with each other. The technical solution of the present disclosure will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0028] Existing house ranking schemes mainly rely on some basic features of houses and user behavior sequences as samples. It is expected that the model will learn the implicit expression relationship between behavior and basic features through these basic features and behavior sequences to estimate the probability of users clicking on houses, and obtain ranking results based on click probability. The learning of this implicit expression relationship cannot be controlled, and the learning effect is poor when there are no behavior sequence samples.

[0029] The embodiments of the present disclosure provide a method for sorting housing sources, which can be implemented on a server. For example, the server responds to a sorting request from a user terminal, executes the method for sorting housing sources provided by the present disclosure, and obtains a sorting order for the housing sources. The server can further generate a sorting result based on the sorting order of each of the multiple housing sources, and return the sorting result to the user terminal.

[0030] Figure 1 FIG. 1 is a schematic diagram showing the overall process of a housing source sorting method M100 according to an embodiment of the present disclosure. Figure 1 The method shown includes steps S110 to S190, wherein the method may be executed by a server device.

[0031] Specifically, Figure 1 The methods shown include: S110 . In response to a housing search request of a target user, determine a plurality of target housing sources.

[0032] The target user refers to the user who initiates the search request. For example, if user A initiates a search request to the server, user A is the target user.

[0033] The target properties are the search results obtained based on the user's search request. For example, when a user enters search conditions on a property search platform to search for properties, the search results obtained are one or more properties that meet the search conditions. At this time, the property search platform needs to sort the one or more properties as search results and display them, so that it can be determined that the target properties to be sorted are one or more properties that meet the search conditions.

[0034] In some embodiments of the present disclosure, in response to a user's search request, the server may return recommended results of the same type in addition to the search results that meet the search criteria. In this case, the results of the same type recommended by the server and the search results that meet the search criteria will be determined together as target properties to be sorted.

[0035] For example, when a property search platform searches for multiple properties that meet the search criteria, it will also find properties that do not meet the search criteria but that the user may prefer based on user behaviors such as the user's historical search history and historical browsing history as recommended properties, and sort the recommended properties together with the properties that meet the search criteria and return them to the user. In this case, both the properties in the search results and the properties recommended by the server are determined as target properties to be sorted.

[0036] S130: Obtain user portrait features of the target user and property attribute features of the target property.

[0037] Among them, user portrait features are features used to represent user attributes, preferences, living habits, behaviors and other information. In some embodiments of the present disclosure, user portrait features may include target user behavior features and target user preference features, which are recorded as user behavior features and user preference features, respectively. In an example of the present disclosure, user portrait features may be stored in the server in the form of text or vector. Features stored in the form of text have the advantage of being easier to find, and features stored in the form of vectors have the advantage of taking up less space and being able to be directly applied to the computing engine.

[0038] User preference features include entities preferred by the target user. An entity refers to something that can be identified as a unique and independent unit. It can be a specific person, thing, or object, or an abstract concept, such as "more", "less", "like", etc. For example, a user preference feature of the target user is "<Business District 1, preference score 0.5>". "Business District 1" and "Preference score 0.5" are the names of the entities preferred by the user and the degree of preference of the user for the entity of that name. In this example, the value range of the preference score is [0, 1].

[0039] User behavior characteristics include entities affected by the behavior of the target user. For example, a user behavior characteristic of the target user is "<Business District 1, searched 5 times in 7 days>". "Business District 1" is the entity affected by the user's search behavior, and "searched 5 times in 7 days" is the user's specific search behavior.

[0040] In one example, the server stores the history of user behavior, and generates the user behavior characteristics and user preference characteristics of the user based on the history of behavior, and the user behavior characteristics and user preference characteristics are saved as the user profile characteristics of the user. When the history of user behavior is updated, the server updates the user preference characteristics according to the specific updated history content.

[0041] The property attribute feature of the target property is used to represent the entity that is the attribute of the target property. For example, a property attribute feature of the target property is "<Property 1, Community A, 2 bedrooms>". "Property 1", "Community A" and "2 bedrooms" are entities representing the name attribute, location attribute and apartment type attribute of the target property respectively.

[0042] In one example, a server is provided with a housing database, which stores multiple searchable housing sources and attribute characteristics of each housing source. When a target housing source is determined, the target housing source can be found among the multiple housing sources included in the housing database, and the attribute characteristics of the target housing source can be obtained.

[0043] S150. Determine entities in user preference features, entities in user behavior features, and entities in property attribute features as target entities in the knowledge graph, and obtain relationship chains between the target entities.

[0044] Among them, the target entity is the entity described by the user preference feature, user behavior feature or property attribute feature. For example, a user behavior feature of the target user is "<Business District 1, searched 5 times in 7 days>", where "searched 5 times in 7 days" is the user's historical behavior, and "Business District 1" is the entity affected by the user's historical behavior, that is, the entity in the user behavior feature, then "Business District 1" is determined as a target entity. For another example, a property attribute feature of the target property is "<Property Source 1, Community A, 2 Bedrooms>", where "Property Source 1", "Community A" and "2 Bedrooms" are all entities in the property attribute feature, then "Property Source 1", "Community A" and "2 Bedrooms" are respectively determined as a target entity.

[0045] The knowledge graph is used to record multiple entities and the relationship between two entities. The relationship chain includes one or more relationships that associate two entities.

[0046] In some embodiments of the present disclosure, the types of association relationships include at least one of a subordinate relationship, a distance relationship, a common category relationship, and a common attribute relationship. Among them, a subordinate relationship indicates that one entity is subordinate to another entity, for example, the subordinate relationship between "Xi'erqi Subway Station" and "Line 13" is "Xi'erqi Subway Station belongs to Line 13". A distance relationship indicates the distance between one entity and another entity, for example, the distance relationship between "House 1" and "School 1" is "within 1000 meters of straight-line distance". A common category relationship indicates that one entity and another entity belong to the same category, for example, the common category relationship between "Second-hand House 1" and "Second-hand House 2" is "Second-hand House". A common attribute relationship indicates that one entity and another entity have the same attributes, for example, the common attribute relationship between the 2-bedroom "House 1" and the 2-bedroom "House 2" is "2 bedrooms".

[0047] Please combine Figure 2 , Figure 2 The following illustrates the knowledge graph and relationship chain in some embodiments of the present disclosure. Figure 2 As shown in the figure, each node in the knowledge graph represents an entity, and the line between two adjacent nodes represents the association relationship between the two nodes. For example, the line between the "Shangdi Dongli" node and the "Yongtaizhuang North Road No. 38 Courtyard" node represents that the two entities "Shangdi Dongli" and "Yongtaizhuang North Road No. 38 Courtyard" are adjacent communities, indicating the distance relationship between "Shangdi Dongli" and "Yongtaizhuang North Road No. 38 Courtyard". For another example, the line between the "Shangdi Dongli" node and the "Shangdi" node represents that the "Shangdi Dongli" entity is subordinate to the "Shangdi" entity at the level of the place to which it belongs, indicating the subordinate relationship between "Shangdi Dongli" and "Shangdi".

[0048] In the knowledge graph, if any two nodes can be connected by an association relationship, then there is a relationship chain between the two nodes, and the relationship chain includes all the associations that connect the two nodes. For example, Figure 2 The "Shangdi Dongli" node and the "Beijing City" node can be connected through the association relationship between the "Shangdi Dongli" node and the "Shangdi" node, the association relationship between the "Shangdi" node and the "Haidian District" node, and the association relationship between the "Haidian District" node and the "Beijing City" node. At this time, the above three associations constitute the relationship chain between the "Shangdi Dongli" node and the "Beijing City" node. According to this relationship chain, it can be determined that the "Shangdi Dongli" entity belongs to the "Beijing City" entity at the level of the place to which it belongs.

[0049] For another example, the "No. 38, Yongtai Zhuang North Road" node and the "Oak Bay Phase I" node can be connected through the association relationship between the "Shangdi Dongli" node and the "No. 38, Yongtai Zhuang North Road" node, and the association relationship between the "Shangdi Dongli" node and the "Oak Bay Phase I" node. At this time, the above two association relationships constitute the relationship chain between the "No. 38, Yongtai Zhuang North Road" node and the "Oak Bay Phase I" node. According to this relationship chain, it can be determined that "No. 38, Yongtai Zhuang North Road" and "Oak Bay Phase I" are both adjacent communities to "Shangdi Dongli".

[0050] For another example, the "Shangdi Dongli" node and the "Shangdi" node can be connected by only one association relationship. In this case, one association relationship between the "Shangdi Dongli" node and the "Shangdi" node constitutes a relationship chain between the "Shangdi Dongli" node and the "Shangdi" node. In other words, a relationship chain can include one, two, three or more association relationships that associate two entities.

[0051] In one example, to search for a relationship chain between two target entities in a knowledge graph, the method may be to determine, based on multiple target entities, whether a relationship chain exists between every two different target entities, and output the relationship chain between the two different target entities if a relationship chain exists.

[0052] S170. Generate house sorting features based on the relationship chain.

[0053] The property ranking feature is a feature used to describe the characteristics of the relationship chain between two target entities. Assuming that there are n types of characteristics of the relationship chain, at most n property ranking features can be generated based on the relationship chain between the two target entities.

[0054] In some embodiments of the present disclosure, it is assumed that there are 4 types of characteristics of the relationship chain, and the types of house sorting features include at least one of binary features, enumeration features, quantity features and vector features, and at most all four. In this case, according to whether the relationship chain between the two target entities has the characteristics of binary features, enumeration features, quantity features and vector features, at most 4 house sorting features can be generated based on the relationship chain between the two target entities. Among them, the binary feature is used to indicate whether there is a relationship chain between the two target entities. The enumeration feature is used to indicate the name of each association relationship that associates the two target entities. The quantity feature is used to indicate the number of association relationships that associate the two target entities. The vector feature is an association relationship in the form of a vector in the relationship chain between the two target entities.

[0055] Please combine Figure 2For example, there is a relationship chain between the "Shangdi Dongli" node and the "Beijing City" node, based on which a binary feature is generated as "<Shangdi Dongli, Beijing City, connectable>". Moreover, the relationship chain between the "Shangdi Dongli" node and the "Beijing City" node includes the association relationship between the "Shangdi Dongli" node and the "Shangdi" node (recorded as subordinate relationship 1), the association relationship between the "Shangdi" node and the "Haidian District" node (recorded as subordinate relationship 2), and the association relationship between the "Haidian District" node and the "Beijing City" node (recorded as subordinate relationship 3), based on which an enumeration feature is generated as "<Shangdi Dongli, Beijing City, subordinate relationship 1, subordinate relationship 2, subordinate relationship 3>". Moreover, the relationship chain between the "Shangdi Dongli" node and the "Beijing City" node includes 3 association relationships, based on which a quantitative feature is generated as "<Shangdi Dongli, Beijing City, 3 association relationships>". Moreover, the relationship chain between the "Shangdi Dongli" node and the "Beijing City" node includes no association relationships in vector form, so no vector feature will be generated.

[0056] Figure 2 The diagram shows a relationship graph in the knowledge graph. A relationship graph can represent a limited number of entities and relationships, so the knowledge graph can represent more entities and relationships by configuring multiple relationship graphs. If a relationship graph is used to represent the relationship between nodes in another relationship graph, the relationship graph is called a subgraph of the other relationship graph, and the other relationship graph is called the main graph. In one example, the information contained in the subgraph can be converted into a vector form through the Embedding technology. In this case, the relationship of a node in the main graph can be a subgraph represented in the form of a vector. If the subgraph represented in the form of a vector is an association relationship in the relationship chain between two target entities, the subgraph represented in the form of a vector is output as a vector-like feature.

[0057] S190: Determine a recommendation value of a target property based on the property ranking feature, and rank the multiple target properties based on the recommendation value.

[0058] In one example, the recommendation value of a target property is calculated based on the property ranking features and the recommendation algorithm. The user portrait features, property attribute features, and property ranking features are input into a neural network model, and the neural network model calculates the input features based on the recommendation algorithm to output the recommendation value of the target property. The neural network model may use the DeepFM (Factorization Machine, abbreviated as FM) model, and calculate the recommendation value of the target property based on the DeepFM algorithm as the recommendation algorithm. Alternatively, a model that uses other recommendation algorithms to calculate the recommendation value may be used, for example, the FFM (Fidld-aware Factorization Machine, abbreviated as FFM) model may be used.

[0059] The recommendation value of the target property indicates the estimated user preference for the target property. The larger the recommendation value of the target property, the higher the probability that the user will select the target property first. Based on this, when the recommendation values ​​of multiple target properties are obtained, the multiple target properties can be sorted in order of the recommendation values ​​to obtain the sorting order of each target property, so that the target property with the larger recommendation value is ranked higher in the sorting order.

[0060] Combined with the previous description, the relationship chain found by step S150 can represent the association relationship between two target entities. At the same time, the target entity is the entity represented in the user portrait feature or the property attribute feature. Based on this, the relationship chain between the two target entities may include: the association relationship between entities covered by different target properties, the association relationship between entities covered by user portrait features and entities covered by target properties, and the association relationship between different entities covered by user portrait features. In other words, the relationship between different target properties, the relationship between user portrait features and target properties, and the relationship between different entities in user portrait features are recorded in an explicit form in the knowledge graph. In this way, based on the relationship chain between two target entities, a property ranking feature is generated, and the property ranking feature is input into the neural network model to calculate the recommendation value. The neural network model can obtain the relationship between different target properties, the relationship between the user portrait feature and the target property, and the relationship between different entities in the user portrait feature from the property ranking feature, so that the neural network model does not need to implicitly learn the expression of the above relationship. On the one hand, it can reduce the difficulty of learning the neural network model, so that the neural network model does not need to learn and discover the expression of the above relationship based on sparse user behavior data. On the other hand, the expression of the relationship in the property ranking feature can be controlled by editing the association relationship in the knowledge graph, so that the relationship expression represented in the property ranking feature has better controllability and interpretability.

[0061] Regarding step S110, in some embodiments of the present disclosure, it may include: Figure 3 Steps S1101 to S1102 are shown.

[0062] S1101: Obtain search conditions in a search request of a target user.

[0063] The search conditions may include search terms, tags, texts, etc. The search conditions may also include screening conditions for filtering out target properties.

[0064] S1102: Search for houses in the house database based on the search conditions, and use the searched houses as target houses.

[0065] The house database stores multiple searchable houses. For example, the house database is a house database, and multiple houses are stored in the house database. The house searched out based on the search condition is the target house.

[0066] In some embodiments of the present disclosure, the user behavior features in the user portrait features include selection times feature, selection sequence feature, search times feature, search sequence feature, and re-search feature. Among them, the selection times feature indicates the number of times a user selects a house within a target time period, and the target time period can usually be set to within 3 days, 5 days, 7 days, etc. The selection sequence feature indicates the sequence composed of the houses selected by the user within the target time period. The search times feature indicates the number of times a user searches for a house within the target time period. The search sequence feature indicates the sequence composed of the houses searched by the user within the target time period. The re-search feature indicates the search time interval between two searches of the same house by the user. The above-mentioned types of user behavior features can, to a certain extent, indicate the user's preference for the selected or searched house. The houses covered in the above-mentioned types of user behavior features are the entities represented in the user behavior features.

[0067] Regarding step S130, in some embodiments of the present disclosure, it may include the following steps: Figure 4 Steps S1301 to S1305 are shown.

[0068] S1301: Obtain historical record information of a target user.

[0069] The historical record information includes at least one of the target user's historical search records for listings and historical click records for listings. The historical search records for listings include information such as the search conditions used by the target user when searching, the time of searching, the listings searched, etc. The historical click records for listings include information such as the listings selected by the target user, the time of selecting the listings, and the number of times the listings were selected.

[0070] S1302. Extract the properties selected by the target user in the target time period and the number of times they were selected from the historical records of clicked properties, generate a selection number feature based on each selected property and the number of times it was selected, and generate a selection sequence feature based on the target time period and each selected property.

[0071] From the historical records of clicked properties, we can extract the properties that the target user has selected during the target time period and the number of times each property has been selected. This extracted information can be used to generate selection count features and selection sequence features.

[0072] For example, the target time period is "recent 3 days". The extracted properties that the target user has selected in the past 3 days include "Property 1" and "Property 2". Among them, "Property 1" was selected once, and "Property 2" was selected twice. Based on the above information, a selection count feature for "Property 1" can be generated: "<Property 1, selected once, recent 3 days>", as well as a selection count feature for "Property 2": "<Property 2, selected twice, recent 3 days>", and a selection sequence feature for "Property 1" and "Property 2": "<Property 1, Property 2, recent 3 days>". Taking the selection count feature of "Property 1": "<Property 1, selected once, recent 3 days>" as an example, the property covered by the selection count feature is "Property 1", and "Property 1" is the entity represented in the selection count feature.

[0073] S1303. Extract the properties searched by the target user in the target time period and the number of times they were searched from the historical records of property searches, generate a search frequency feature based on each searched property and the number of times it was searched, and generate a search sequence feature based on the target time period and each searched property.

[0074] In the historical search records of the historical record information, multiple listings searched by the target user in the target time period and the number of times each listing was searched can be extracted. The extracted information can be used to generate search frequency features and search sequence features.

[0075] For example, the target time period is "recent 3 days". The extracted properties that the target user has searched for in the past 3 days include "Property 1" and "Property 2". Among them, "Property 1" has been searched once, and "Property 2" has been searched twice. Based on the above information, a search count feature for "Property 1" can be generated: "<Property 1, searched 1 time, recent 3 days>", as well as a search count feature for "Property 2": "<Property 2, searched 2 times, recent 3 days>", and a search sequence feature for "Property 1" and "Property 2": "<Property 1, Property 2, recent 3 days>". Taking the search count feature of "Property 2": "<Property 2, searched 2 times, recent 3 days>" as an example, the property covered by the search count feature is "Property 2", and "Property 2" is the entity represented in the search count feature.

[0076] S1304: extract the time interval between two searches of the target user for the same property from the historical property search records, and generate a re-search feature based on the same property searched twice and the time interval.

[0077] In the historical search records of the historical record information, the same house that the target user has searched twice and the respective search times of the two searches can be extracted. The time interval between the target user's two searches for the same house can be determined based on the respective search times of the two searches, so that the same house that has been searched twice and the time interval between the two searches can be used to generate a re-search feature.

[0078] For example, the time interval between the two searches of "House 1" by the target user is extracted to be 1 day. Based on the above information, a re-search feature for "House 1" can be generated: "<House 2, 1 day interval>". "House 1" is the entity represented in the re-search feature.

[0079] S1305: Obtain the target user's preference score for the search condition based on the historical records of house search, and generate a user preference feature based on a search condition and its preference score.

[0080] The target user's preference score for the search condition is used to indicate the user's preference for the search condition. In one example, the frequency of occurrence of a search condition in the historical record information can be calculated, and the frequency can be normalized to obtain a normalized value, and the obtained normalized value is used as the preference score of the search condition.

[0081] For example, one of the search conditions in the historical search records for housing sources is "urban area 1", and the frequency of "urban area 1" in the historical record information is 1 time for every 20 search conditions, and the normalized value obtained by normalization is 0.41. Based on the above information, the user preference feature for "urban area 1" can be generated as: "<urban area 1, preference score 0.41>".

[0082] In one example, the search condition covered by the user preference feature is the entity represented in the user preference feature. For example, in the user preference feature "<Urban District 1, preference score 0.41>" about "Urban District 1", the search condition "Urban District 1" is the entity represented in the user preference feature.

[0083] In some embodiments of the present disclosure, the house sorting method M100 further includes: Figure 5 Step S160 is shown.

[0084] S160. In response to an entity represented in the user portrait feature not being found in the knowledge graph, the entity that cannot be found is stored in the knowledge graph as a newly added entity, and the association relationship between the newly added entity and other entities in the knowledge graph is recorded in the knowledge graph.

[0085] Step S160 occurs after step S150. When an entity represented in the user portrait feature cannot be found in the knowledge graph, it may be because the entity is not recorded in the knowledge graph. In this case, step S160 can be used to record the entity and its corresponding association in the knowledge graph.

[0086] In one example, storing a new entity into a knowledge graph includes creating a new node in the knowledge graph to represent the new entity, and generating connections between the node representing the new entity and other nodes based on the association relationship between the new entity and existing entities in the knowledge graph.

[0087] Regarding step S160, in some embodiments of the present disclosure, the association relationship between the newly added entity and other entities in the knowledge graph can be achieved by Figure 6 The acquisition is performed in steps S1601 to S1605 shown.

[0088] S1601. Query in the knowledge graph whether the newly added entity is an attribute of another entity in the knowledge graph. When the newly added entity is an attribute of another entity in the knowledge graph, generate a subordinate relationship corresponding to the newly added entity based on the newly added entity and the other entity in the queried knowledge graph.

[0089] For example, the newly added entity is "Xi'erqi Subway Station", and the attributes of another entity "Line 13" in the knowledge graph include Xi'erqi Subway Station. In this case, based on the newly added entity "Xi'erqi Subway Station" and another entity "Line 13" in the knowledge graph, a subordinate relationship "Xi'erqi Subway Station belongs to Line 13" can be generated.

[0090] S1602. Obtain the attributes of the newly added entity, and query in the knowledge graph whether the attributes of the newly added entity are another entity in the knowledge graph. When the attributes of the newly added entity are another entity in the knowledge graph, generate a subordinate relationship corresponding to the newly added entity based on the newly added entity and the other entity in the queried knowledge graph.

[0091] For example, the newly added entity is "Line 13", the attributes of "Line 13" include Xi'erqi Subway Station, and another entity in the knowledge graph is "Xi'erqi Subway Station". In this case, based on the newly added entity "Line 13" and another entity "Xi'erqi Subway Station" in the knowledge graph, a subordinate relationship "Xi'erqi Subway Station belongs to Line 13" can be generated.

[0092] S1603. Obtain the distance between the newly added entity and another entity in the knowledge graph, and when the obtained distance is less than a distance threshold, generate a distance relationship corresponding to the newly added entity based on the obtained distance.

[0093] For example, if the distance threshold is within 1000 meters in straight line distance, if the distance between one entity and another entity is less than the distance threshold, the two entities are considered to be adjacent. Among them, both entities are buildings. Assume that the newly added entity is "Business District 1", another entity in the knowledge graph is "House 1", and the straight line distance between Business District 1 and House 1 is 800 meters. In this case, based on the newly added entity "Business District 1" and another entity "House 1" in the knowledge graph, a distance relationship "Business District 1 is adjacent to House 1" can be generated.

[0094] S1604. Obtain the category of the newly added entity, and query entities of the same category in the knowledge graph. When entities of the same category are found, generate a common category relationship corresponding to the newly added entity based on the category of the newly added entity.

[0095] For example, the newly added entity is "Second-hand House Source 2", the category of "Second-hand House Source 2" is second-hand house source, and the entity of the second-hand house source category found in the knowledge graph also includes "Second-hand House Source 1". In this case, based on the newly added entity "Second-hand House Source 2" and another entity "Second-hand House Source 1" in the knowledge graph, a common category relationship "Second-hand House Source 2 and Second-hand House Source 1 are both second-hand houses" can be generated.

[0096] S1605. Query the knowledge graph for entities with the same attributes as the newly added entity. When entities with the same attributes are found, generate common attribute relationships corresponding to the newly added entity based on the attributes of the newly added entity.

[0097] For example, the newly added entity is "House 2", one of the attributes of "House 2" is "2 bedrooms", and the entity with the attribute "2 bedrooms" is found in the knowledge graph, which also includes "House 1". In this case, based on the newly added entity "House 2" and another entity "House 1" in the knowledge graph, a common attribute relationship "House 2 and House 1 are both adjacent and have 2 bedrooms" can be generated.

[0098] Regarding step S170, in some embodiments of the present disclosure, it may include the following steps: Figure 7 Steps S1701 to S1704 are shown.

[0099] S1701. Query whether there is a relationship chain between two target entities, obtain an existence query result, and generate a binary class feature based on the existence query result.

[0100] Please combine Figure 2For example, if a relationship chain is found between the "Shangdi Dongli" node and the "Beijing City" node, a binary class feature "<Shangdi Dongli, Beijing City, connectable>" is generated based on this. Assuming that there is no relationship chain between the "Shangdi Dongli" node and the "Beijing City" node, the corresponding binary class feature generated is "<Shangdi Dongli, Beijing City, not connected>".

[0101] S1702: Obtain the names of each association relationship included in the relationship chain between the two target entities, and generate an enumeration class feature based on the obtained names, where the enumeration class feature is used to represent the names of each association relationship that associates the two target entities.

[0102] Please combine Figure 2 For example, the relationship chain between the "Shangdi Dongli" node and the "Beijing City" node includes the association relationship between the "Shangdi Dongli" node and the "Shangdi" node (recorded as subordinate relationship 1), the association relationship between the "Shangdi" node and the "Haidian District" node (recorded as subordinate relationship 2), and the association relationship between the "Haidian District" node and the "Beijing City" node (recorded as subordinate relationship 3). Based on this, an enumeration class feature is generated as "<Shangdi Dongli, Beijing City, subordinate relationship 1, subordinate relationship 2, subordinate relationship 3>".

[0103] S1703: Count the number of association relationships included in the relationship chain between the two target entities, and generate a quantity feature based on the counted number, where the quantity feature is used to represent the number of association relationships that associate the two target entities.

[0104] Please combine Figure 2 For example, the relationship chain between the "Shangdi Dongli" node and the "Beijing City" node includes 3 association relationships, and a quantitative feature "<Shangdi Dongli, Beijing City, 3 association relationships>" is generated based on this.

[0105] S1704. Query the vector data in each association relationship included in the relationship chain between two target entities, and use the queried vector data as a vector class feature.

[0106] Please combine Figure 2 For example, in the association relationship included in the relationship chain between the "Shangdi Dongli" node and the "Beijing City" node, the association relationship between the "Shangdi Dongli" node and the "Shangdi" node is a knowledge graph subgraph in vector form, then the vector representing the association relationship between the "Shangdi Dongli" node and the "Shangdi" node is taken as a vector-like feature.

[0107] Regarding step S190, in some embodiments of the present disclosure, it may include the following steps: Figure 8 Steps S1901 to S1902 are shown.

[0108] S1901, transmitting user portrait features, property attribute features of multiple target properties, and property ranking features to a recommendation module.

[0109] The recommendation module is used to determine the target property from the input features, and to calculate the recommendation value of the input features based on the recommendation algorithm to obtain the recommendation value of the target property. The recommendation calculation module can be a DeepFM model, or other models that use the recommendation algorithm for calculation, such as the FFM model.

[0110] S1902. Obtain recommendation values ​​of multiple target properties returned by the recommendation module.

[0111] When a traditional model uses a recommendation algorithm to calculate a recommendation value, it is necessary to pre-train the model to learn to use the user behavior characteristics and the characteristics of the house to be recommended to construct features that can represent the correlation between the user behavior and the house to be recommended. The learning difficulty is large and the effect is poor. In the embodiment of the present disclosure, the house sorting feature can be generated based on the relationship chain in the knowledge graph. The house sorting feature can represent the correlation between the user portrait feature and the target house to be sorted, the correlation between different entities represented in the user portrait feature, and the correlation between different target houses. When calculating the recommendation value, the house sorting feature representing the above relationship can be provided to the recommendation module to calculate the recommendation value, so that the recommendation module can obtain a relatively accurate correlation relationship without learning to construct the above relationship, so that the recommendation value output by the recommendation module has higher reliability.

[0112] Based on any of the above embodiments, the present disclosure also provides a housing source sorting device. Fig. 9 It is a schematic block diagram of the structure of a housing source sorting device according to an embodiment of the present invention.

[0113] like Fig. 9 As shown, the house source sorting device includes: The response module 110 is used to respond to the housing search request of the target user and determine multiple target housing sources.

[0114] The acquisition module 120 is used to acquire the user portrait characteristics of the target user and the property attribute characteristics of the target property.

[0115] The search module 130 is used to determine entities in user preference features, entities in user behavior features, and entities in property attribute features as target entities in the knowledge graph, and obtain relationship chains between the target entities.

[0116] The feature generation module 140 is used to generate house sorting features based on the relationship chain.

[0117] The sorting module 150 is used to determine the recommendation value of the target house based on the house sorting feature, and sort the multiple target houses based on the recommendation value.

[0118] The implementation process of the functions and effects of each module in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, which will not be repeated here.

[0119] The executor of the housing source sorting method in the specific implementation manner of the present disclosure may be an electronic device such as a server device.

[0120] Therefore, based on any of the above-mentioned embodiments, the present disclosure also provides an electronic device, which can execute the housing source sorting method of any of the embodiments described above in the present disclosure, and the housing source sorting device of any of the embodiments described above can be configured on the electronic device (for example, configured in the form of software on the memory of the electronic device).

[0121] Fig.10 It is a schematic block diagram of the structure of an electronic device 1000 equipped with a house ranking device according to an embodiment of the present disclosure.

[0122] The hardware structure of the electronic device 1000 can be implemented using a bus architecture. The bus architecture can include any number of interconnected buses and bridges, depending on the specific application and overall design constraints of the hardware. The bus 1100 connects various circuits including one or more processors 1200, memory 1300, and / or hardware modules together. The bus 1100 can also connect various other circuits 1400 such as peripherals, voltage regulators, power management circuits, external antennas, etc.

[0123] The bus 1100 may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Component (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the figure only uses one connecting line, but does not mean that there is only one bus or one type of bus.

[0124] The present disclosure also provides a readable storage medium, in which a computer program is stored, and the computer program is used to implement the above method when executed by a processor. "Readable storage medium" can be any device that can contain, store, communicate, propagate or transmit a program for use in an instruction execution system, device or equipment or in combination with these instruction execution systems, devices or equipment. More specific examples of readable storage media include the following: an electrical connection portion (electronic device) with one or more wirings, a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and editable read-only memory (EPROM or flash memory), an optical fiber device, and a portable read-only memory (CDROM), etc.

[0125] The present disclosure also provides a computer program product. The method of the present disclosure can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instruction is loaded and executed, the process or function of the present disclosure is executed in whole or in part.

[0126] A computer program or instruction may be stored in a readable storage medium or transmitted from one readable storage medium to another readable storage medium, for example, a computer program or instruction may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired or wireless means. The readable storage medium may be any available medium that can be accessed or a data storage device such as a server, data center, etc. that integrates one or more available media. The available medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; it may also be an optical medium, such as a digital video disk; it may also be a semiconductor medium, such as a solid state drive. The computer readable storage medium may be a volatile or non-volatile storage medium, or may include both volatile and non-volatile types of storage media.

[0127] Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as methods, systems, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0128] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the present disclosure. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks 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 sequencing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable sequencing device generate instructions for implementing the processes in the flowchart and / or block diagram. 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.

[0129] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable sequencing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0130] These computer program instructions may also be loaded onto a computer or other programmable sequencing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide for implementing the process described in the flow chart. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0131] In the description of this specification, the description with reference to the terms "one embodiment / method", "some embodiments / methods", "example", "specific example", or "some examples" etc. means that the specific features, structures, or characteristics described in conjunction with the embodiment / method or example are included in at least one embodiment / method or example of the present disclosure. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment / method or example. Moreover, the specific features, structures, or characteristics described may be combined in any one or more embodiments / methods or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments / methods or examples described in this specification and features of different embodiments / methods or examples, unless they are contradictory.

[0132] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present disclosure, "plurality" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0133] Those skilled in the art should understand that the above embodiments are only for the purpose of clearly illustrating the present disclosure, and are not intended to limit the scope of the present disclosure. For those skilled in the art, other changes or modifications may be made based on the above disclosure, and these changes or modifications are still within the scope of the present disclosure.

[0134] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the type, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0135] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0136] As an optional but non-limiting implementation, in response to receiving an active request from the user, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0137] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet the relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0138] At the same time, it is understandable that the data involved in the technical solution of the present disclosure (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and relevant provisions.

Claims

1. A method for sorting housing sources, characterized in that: include: In response to a target user's housing search request, determining a plurality of target housing sources; Obtaining user portrait features of the target user and property attribute features of the target property, wherein the user portrait features include user preference features and user behavior features, wherein the user preference features include entities preferred by the target user, the user behavior features include entities affected by the historical behavior of the target user, and the property attribute features include entities representing the properties of the property; Determine in the knowledge graph the entities in the user preference feature, the entities in the user behavior feature, and the entities in the property attribute feature as target entities, and obtain the relationship chain between the target entities; Generate a property sorting feature based on the relationship chain; as well as The recommendation value of the target house is determined based on the house ranking feature, and the multiple target houses are ranked based on the recommendation value.

2. The housing source sorting method according to claim 1, characterized in that: The type of the house source sorting feature includes at least one of a binary feature, an enumeration feature, a quantity feature, and a vector feature; generating the house source sorting feature based on the relationship chain includes: Query whether there is a relationship chain between the two target entities to obtain an existence query result, and generate the binary feature based on the existence query result, wherein the binary feature is used to indicate whether there is a relationship chain between the two target entities; and / or Obtaining the names of each association relationship included in the relationship chain between the two target entities, and generating the enumeration class feature based on the acquired names, wherein the enumeration class feature is used to represent the names of each association relationship that associates the two target entities; and / or Counting the number of each association relationship included in the relationship chain between the two target entities, and generating the quantity feature based on the counted number, wherein the quantity feature is used to represent the number of association relationships that associate the two target entities; and / or The vector data in each association relationship included in the relationship chain between the two target entities is queried, and the queried vector data is used as the vector class feature.

3. The housing source sorting method according to claim 1, characterized in that: Also includes: In response to an entity represented in the user portrait not being found in the knowledge graph, the entity that cannot be found is stored in the knowledge graph as a newly added entity, and the association relationship between the newly added entity and other entities in the knowledge graph is recorded in the knowledge graph.

4. The housing source sorting method according to claim 3, characterized in that: The type of the association relationship includes at least one of a subordinate relationship, a distance relationship, a common category relationship, and a common attribute relationship. The association relationship between the newly added entity and other entities in the knowledge graph is recorded in the knowledge graph, including: Querying in the knowledge graph whether the newly added entity is an attribute of another entity in the knowledge graph, and when the newly added entity is an attribute of another entity in the knowledge graph, generating the subordinate relationship corresponding to the newly added entity based on the newly added entity and the queried another entity in the knowledge graph; and / or Obtaining the attributes of the newly added entity, and querying in the knowledge graph whether the attributes of the newly added entity are another entity in the knowledge graph, and when the attributes of the newly added entity are another entity in the knowledge graph, generating the subordinate relationship corresponding to the newly added entity based on the newly added entity and the queried another entity in the knowledge graph; and / or Obtaining a distance between the newly added entity and another entity in the knowledge graph, and when the obtained distance is less than a distance threshold, generating the distance relationship corresponding to the newly added entity based on the obtained distance; and / or Obtaining the category of the newly added entity, and searching for entities of the same category in the knowledge graph, and when entities of the same category are found, generating the common category relationship corresponding to the newly added entity based on the category of the newly added entity; and / or Entities having the same attributes as the newly added entity are queried in the knowledge graph, and when the entities having the same attributes are found, the common attribute relationship corresponding to the newly added entity is generated based on the attributes of the newly added entity.

5. The housing source sorting method according to claim 1, characterized in that: The entity in the user behavior feature includes a house source, the type of the user behavior feature includes at least one of a selection number feature, a selection sequence feature, a search number feature, a search sequence feature, and a re-search feature, and the entity in the user preference feature includes a search condition; obtaining the user portrait feature of the target user includes: Acquire the historical record information of the target user, wherein the historical record information includes at least one of the target user's historical search records for housing resources and historical click records for housing resources; Extracting the properties selected by the target user in the target time period and the number of times they were selected from the records of the historically clicked properties, generating a selection number feature based on each selected property and the number of times it was selected, and generating a selection sequence feature based on the target time period and each selected property; and / or Extracting the properties searched by the target user in the target time period and the number of times they were searched from the records of the historical property searches, generating a search number feature based on each searched property and the number of times it was searched, and generating a search sequence feature based on the target time period and each searched property; and / or Extracting the time interval between two searches of the same property by the target user from the historical property search records, and generating the re-search feature based on the two searches of the same property and the time interval; and / or The target user's preference score for the search condition is obtained based on the historical records of house search, and a user preference feature is generated based on the search condition and its preference score.

6. The housing source sorting method according to claim 1, characterized in that: Determining the recommendation value of the target property based on the property ranking feature includes: Transmitting the user portrait features, the property attribute features of the plurality of target properties, and the property ranking features to a recommendation module; and Obtain recommendation values ​​of the multiple target properties returned by the recommendation module.

7. The housing source sorting method according to claim 1, characterized in that: Identify multiple target properties, including: Obtaining search conditions in the search request of the target user; and The house in the house database is searched based on the search condition, and the searched house is determined as the target house.

8. An electronic device, characterized in that: include: A memory storing execution instructions; as well as A processor, wherein the processor executes the execution instructions stored in the memory, so that the processor executes the housing source sorting method according to any one of claims 1 to 7.

9. A readable storage medium, characterized in that: The readable storage medium stores execution instructions, which, when executed by a processor, are used to implement the housing source sorting method according to any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the housing source sorting method described in any one of claims 1 to 7 is implemented.

Images