A method for intelligent recommendation of apartment types

Abstract

An embodiment of the present invention provides a method for intelligently recommending apartment types, comprising: receiving a current apartment type input by a user; obtaining a first candidate apartment type similar to the current apartment type based on the current apartment type; and generating a first list of recommended apartment types based on the first candidate apartment types. This method utilizes big data statistics and machine learning to continuously estimate the user's living needs as they draw their apartment type, thereby achieving real-time intelligent apartment type recommendations.

Description

Technical Field

[0001] An embodiment of the present invention relates to a method for intelligently recommending apartment types. Background Art

[0002] In the real estate sector, as consumers' demand for higher quality living continues to grow, they're increasingly focused on the home itself, with apartment layout being one of the most important attributes influencing the living experience. To meet the needs of consumers with varying demographics and living scenarios, apartment layouts are increasingly differentiated, such as the H-shaped and airplane-shaped layouts. Each of these iconic layouts addresses the living needs and scenarios of a specific consumer. For example, the airplane-shaped layout addresses the intergenerational differences in living arrangements for three generations living together. With bedrooms located at opposite ends of the living room, this design allows for shared living while minimizing interference.

[0003] However, when users express such needs, it's often difficult to abstractly express them using a specific apartment type. A more intuitive approach is to sketch out their desired apartment on a canvas. The problem lies in the following: First, drawing an apartment requires a high level of professionalism. Existing apartment drawing products are almost always used by professionals like designers and architects, leaving ordinary users without the tools to express their apartment needs. Second, users may only be able to identify a subset of the apartment's features, without fully understanding the final apartment that meets their needs. Existing apartment recommendations can't quickly understand a user's living needs based on these partial features.

[0004] Therefore, there is a need for a method and device for intelligently recommending apartment types that can overcome the above-mentioned problems, so that users can express their needs for apartment types and understand and analyze the needs in real time to achieve intelligent recommendation of apartment types.

[0005] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present invention, and therefore it may contain information that does not constitute the prior art known to those skilled in the art. Summary of the Invention

[0006] In order to solve the problems existing in the prior art, the present invention provides an intelligent apartment type recommendation method and device, which allows users to express their needs for apartment types through simple and easy-to-use apartment type editing and abstract tools, and adopts big data statistics and machine learning methods to guess the user's living needs in real time during the process of the user drawing the apartment type, thereby overcoming the problems existing in the prior art.

[0007] The present invention provides an intelligent apartment type recommendation method, comprising: a receiving step of receiving a current apartment type input by a user; an acquiring step of acquiring a first candidate apartment type similar to the current apartment type based on the current apartment type; and a generating step of generating a first recommended apartment type list based on the first candidate apartment type.

[0008] According to one embodiment of the present invention, the intelligent apartment type recommendation method also includes: extracting the layout features of the current apartment type, and obtaining a first alternative apartment type similar to the current apartment type based on the current apartment type, including: comparing the layout features of the current apartment type with the layout features of the apartments in the apartment type library, and searching for an apartment type similar to the current apartment type as the first alternative apartment type.

[0009] According to one embodiment of the present invention, the current apartment type is edited by the user using a preset apartment type editing tool. The preset apartment type editing tool has a user operation interface, which includes a drawing area and an element area. The user edits the apartment type by drawing in the drawing area using apartment type elements in the element area. The apartment type elements include different types of rooms.

[0010] According to an embodiment of the present invention, the intelligent apartment type recommendation method further includes: in response to detecting a modification to the current apartment type, re-performing the obtaining step and the generating step on the modified current apartment type.

[0011] According to an embodiment of the present invention, the modification of the current apartment layout includes one or more of the following: adding, reducing, moving, rotating, and scaling apartment layout elements.

[0012] According to one embodiment of the present invention, the apartment type library includes multiple apartment type clusters, each apartment type cluster has a central feature vector; by comparing the layout features of the current apartment type with the layout features of the apartment types in the apartment type library, searching for an apartment type similar to the current apartment type as the first alternative apartment type includes: quantizing the layout feature vectors of the apartment types in the apartment type library and the current apartment type, respectively, to obtain the feature vector of each apartment type in the apartment type library and the feature vector of the current apartment type; calculating the distance between the current apartment type feature vector and the central feature vector of each apartment type cluster to obtain the cluster distance; extracting the apartment type clusters whose cluster distance is less than a first threshold as similar apartment type clusters; and taking the apartment type under the similar apartment type cluster as the first alternative apartment type.

[0013] According to one embodiment of the present invention, generating a first recommended apartment type list based on the first alternative apartment type includes: obtaining a user portrait of the user, and vectorizing the user portrait to obtain a user portrait feature vector; respectively calculating the distance between the feature vector of each apartment type in the first alternative apartment type and the current apartment type feature vector and the user portrait feature vector to obtain an apartment type similarity distance and a user portrait distance; respectively setting weights for the apartment type similarity distance and the user portrait distance, and calculating the weighted values of the two to obtain a total similarity distance; and extracting apartment types whose total similarity distance is less than a second threshold as the first recommended apartment type list.

[0014] According to another aspect of the present invention, an intelligent apartment type recommendation device is also provided, including: a receiving module, which is configured to receive a current apartment type input by a user; an acquisition module, which is configured to acquire a first alternative apartment type similar to the current apartment type based on the current apartment type; and a generation module, which is configured to generate a first recommended apartment type list based on the first alternative apartment type.

[0015] According to one embodiment of the present invention, the intelligent apartment type recommendation device also includes: an update module, which is configured to: monitor whether the current apartment type edited by the user has changed, and if so, use the changed apartment type as the current apartment type; if not, continue to monitor whether the apartment type edited by the user has changed.

[0016] According to another aspect of the present invention, a computer device is also provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the above-described intelligent apartment recommendation method are implemented.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the steps of the above-mentioned intelligent apartment type recommendation method are implemented.

[0018] According to another aspect of the present invention, a computer program product is provided, comprising computer instructions, which, when executed by a processor, implement the steps of the above-mentioned method for intelligently recommending apartment types.

[0019] The intelligent apartment type recommendation method and device provided by the present invention adopt big data statistics and machine learning methods to continuously guess the user's living needs while the user is drawing the apartment type, thereby realizing real-time intelligent apartment type recommendation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and other features of the present invention will be described in detail below with reference to certain exemplary embodiments shown in the accompanying drawings, which are given below by way of illustration only and therefore do not limit the present invention, in which:

[0021] Figure 1 An exemplary system architecture is shown that is applicable to a specific implementation of the intelligent apartment type recommendation method of the present invention.

[0022] Figure 2 A flowchart of a method for intelligently recommending apartment types according to an embodiment of the present invention is shown.

[0023] Figure 3 Another embodiment of the present invention is shown. Figure 2 Flowchart of the intelligent apartment recommendation method with the same steps in .

[0024] Figure 4 The user interface of the apartment editing tool according to one embodiment of the present invention is shown.

[0025] Figure 5 An embodiment of the present invention is shown Figure 1 Flowchart of the process of step S1200.

[0026] Figure 6 An embodiment of the present invention is shown Figure 1 Flowchart of the process of step S1300.

[0027] Figure 7 A schematic structural diagram of an intelligent apartment type recommendation device according to an embodiment of the present invention is shown.

[0028] Figure 8 A schematic structural diagram of an intelligent apartment type recommendation device according to another embodiment of the present invention is shown. DETAILED DESCRIPTION

[0029] The present invention is described in detail below through specific embodiments so that those skilled in the art can easily implement the present invention according to the contents disclosed in this specification. The embodiments described below are only some embodiments of the present invention, not all. Based on the embodiments described in this specification, all other embodiments obtained by those skilled in the art without paying creative work are within the scope of protection of the present invention. It should be noted that the embodiments in this specification and the features in the embodiments can be combined with each other unless there is a conflict.

[0030] The terms used herein are only for the purpose of illustrating specific embodiments and are not intended to limit the present invention. Unless the context otherwise specifically indicates, the singular forms "one", "a kind of" and "the" used herein also include the plural forms "multiple", "multiple" and "the". The words "first", "second" and the like used herein are only used to distinguish different features, steps, operations, elements and / or parts, and neither represent any specific technical meaning nor the necessary logical order between them. The word "multiple" used herein may refer to two or more, and the word "at least one" may refer to one, two or more. For any feature, step, operation, element and / or part mentioned herein, unless the context otherwise specifically indicates, it can generally be understood as one or more. It should also be understood that the words "include" and / or "comprising" used herein refer to the presence of the features, steps, operations, elements and / or parts, but do not exclude the presence or increase of one or more other features, steps, operations, elements, parts and / or their collections. The word "and / or" used herein includes any or all combinations of one or more related listed items. The element suffixes “module” and “unit” herein are only used for convenience of description and, thus, can be used interchangeably without having any distinct meanings or functions.

[0031] When prior art related to the description of the present invention is obvious to those skilled in the art, detailed descriptions thereof will be omitted. It should also be understood that the descriptions of the various embodiments in this specification focus on the differences between the embodiments. The similarities or similarities between the embodiments can be referenced to each other, and for the sake of brevity, this specification does not provide a detailed description of each one.

[0032] like Figure 1 The schematic diagram shows an exemplary system architecture 100 applicable to a specific embodiment of the intelligent apartment recommendation method of the present invention. System architecture 100 may include terminal devices 101, 102, and 103, as well as a network 104 and a server 105. Network 104 is used to provide communication between terminal devices 101, 102, and 103 and server 105, and may include various connection types, such as wired or wireless communication or fiber optic cables.

[0033] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104. Various communication client applications can be installed on terminal devices 101, 102, and 103, such as image and video capture applications, text input applications, web browser applications, professional field application software, search applications, instant messaging tools, email clients, social platform software, etc.

[0034] In a specific implementation, the terminal devices 101, 102, 103 can be implemented as hardware or software according to actual needs. When the terminal devices 101, 102, 103 are implemented as hardware, they can be various electronic devices with a (touch) display screen and support various inputs such as voice and text, including but not limited to personal computers (including laptops and desktop computers), tablet computers, smart phones, car terminals, e-book readers, video players, etc. When the terminal devices 101, 102, 103 are implemented as software, they can be installed in a suitable electronic device and implemented as multiple software or software modules (for example, to provide distributed services), or they can be implemented as a single software or software module. It should be understood that Figure 1 The examples of terminal devices 101, 102, and 103 described in and above are merely examples and should not be understood as specific limitations.

[0035] Server 105 may be a server that provides various services, such as a backend server that analyzes, responds to, and supports various information input by terminal devices 101, 102, and 103, such as control signals, voice, or text information. The backend server may analyze and process received control signals, voice, or target text information, and feed the processing results back to terminal devices 101, 102, and 103 via network 104.

[0036] In specific implementations, the server 105 can be implemented as hardware or software according to actual needs. When the server 105 is implemented as hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or it can be implemented as a single server. When the server 105 is implemented as software, it can be implemented as multiple software or software modules (for example, to provide distributed services), or it can be implemented as a single software or software module. It should be understood that Figure 1 The server 105 examples described in and above are provided herein only as examples and should not be construed as being particularly limiting.

[0037] It should be noted that the intelligent apartment type recommendation method provided in the embodiments of the present application can be executed by the terminal devices 101, 102, and 103, or by the server 105, or by the terminal devices 101, 102, and 103 in conjunction with the server 105. Accordingly, the intelligent apartment type recommendation device can be provided in the terminal devices 101, 102, and 103, or in the server 105, or in both the terminal devices 101, 102, and 103 and the server 105.

[0038] It is understandable that when the intelligent apartment type recommendation method provided in the embodiment of the present application is executed by the terminal devices 101 , 102 , and 103 , the above-mentioned system architecture 100 may not include the network 104 and the server 105 .

[0039] It should be understood that Figure 1 The number and types of terminal devices, networks and servers in the embodiment are merely illustrative. In a specific implementation, any number and type of terminal devices, networks and servers may be used according to actual needs.

[0040] like Figure 2 As shown, the present invention provides a method 1000 for intelligently recommending apartment types. Specifically, method 1000 includes a receiving step at S1100 of receiving a current apartment type input by a user; an obtaining step at S1200 of obtaining a first candidate apartment type similar to the current apartment type based on the current apartment type; and a generating step at S1300 of generating a first recommended apartment type list based on the first candidate apartment types.

[0041] Specifically, in Figure 2 In a preferred embodiment of the present invention of step S1100 shown in , the current apartment type is edited by the user using a preset apartment type editing tool, and the preset apartment type editing tool has a user operation interface, and the user operation interface includes a drawing area and an element area, wherein the user edits the apartment type by drawing apartment type elements in the element area in the drawing area, and the apartment type elements include different types of rooms. Specifically, the apartment type editing tool can be composed of multiple modules such as a database, a server, and a front end. Those skilled in the art can easily understand that the front end implemented as a user operation interface can display information to the user by calling data in the database and performing logical processing through the server, that is, the user can use the functions of the apartment type editing tool on the user operation interface and obtain corresponding information.

[0042] See further Figure 4 , which schematically shows the user interface of the apartment editing tool according to an embodiment of the present invention. Figure 4 As shown, the operation interface is mainly divided into three parts: element area, drawing area and apartment type recommendation area. The element area can be set on the right side of the operation interface, which enumerates common apartment elements such as living rooms or partitions, such as bedrooms, living rooms / dining rooms, kitchens, bathrooms, balconies, etc. Users can move the corresponding partition blocks to the drawing area by clicking and dragging their fingers, thus completing the addition of a partition. The drawing area is the area where the current apartment layout is presented. After dragging apartment elements to this area, users can complete editing operations such as adding, deleting, moving, rotating, and scaling apartment elements in this area. Figure 4The apartment type recommendation area is located at the bottom of the operation interface. The apartment type recommendation area is used to display the apartment types recommended to the user so that the user can choose the apartment type they like and prepare for the subsequent output of the corresponding housing to the user. In the embodiment, various editing operations can be set to be implemented by different gestures as needed, for example, zooming in and out with two fingers, deleting with double-clicking the finger joints, etc. Alternatively, the above-mentioned area can also be set at other locations in the user interface as needed, wherein the apartment type recommendation area may not be set in the same interface as the drawing area and the element area, but after the user completes the apartment type editing, the recommended apartment type is displayed to the user in the next interface. Furthermore, the apartment type editing tool can be installed in Figure 1 On the terminal devices 101, 102, and 103 shown, the server 105 communicates with the terminal devices 101, 102, and 103 in real time. Through the sufficiently simplified interface of the apartment editing tool, the user is allowed to express the apartment layout of his dreams in his mind anytime and anywhere, thereby facilitating the discovery and understanding of the user's more specific apartment needs.

[0043] Alternatively, the user can also input the current apartment type through other methods such as voice and text, and the preset apartment type editing tool can also be created to draw the apartment type through voice, handwriting, click options, etc. The above embodiments are only for illustrative purposes and are not intended to limit the present invention.

[0044] like Figure 3 As shown, it shows some steps of the flowchart of the intelligent apartment type recommendation method 2000 according to another embodiment of the present invention. Specifically: at S2010, a user portrait of the user is obtained, and the user portrait includes user features related to the user; at S2020, the apartment type corresponding to the user portrait is searched in the apartment type library as the second alternative apartment type; at S2030, a second recommended apartment type list is generated based on the second alternative apartment type; at S2040, it is determined whether the user has selected any apartment type in the second recommended apartment type list: if yes, the method proceeds to S2050 to output the house corresponding to the apartment type selected by the user; if not, the method proceeds to S2060 to determine whether the apartment type edited by the user has changed: if yes, the method proceeds to S2070 to use the changed apartment type as the current apartment type; if not, the method repeats the step of determining whether the apartment type edited by the user has changed in S2060 (i.e., monitoring whether the current apartment type edited by the user has changed). After step S2070, the same Figure 2 The same steps S1100 to S1300 as shown in FIG. Figure 3 This part is omitted.

[0045] Specifically, in Figure 3At step S2010 shown, obtaining the user portrait of the user can be initiated based on recognizing the user's actions of logging into the housing website, opening the apartment editing tool, and searching for housing. For example, when the user opens the apartment editing tool through the terminal devices 101, 102, and 103 and wants to edit the ideal apartment, the server 105 can initiate the acquisition of the user portrait by recognizing this action, thereby laying the foundation for subsequent assistance in obtaining recommended housing for the user. In the field of real estate, user portraits can be obtained, for example, based on the user's operation log and the housing knowledge graph. Based on the user portraits, the user's needs for housing types can be understood more deeply and accurately. As for the specific steps of obtaining user portraits, those skilled in the art can obtain them based on known methods as needed, and will not be repeated here.

[0046] Later, in Figure 3 At step S2020, a unit library is searched for a unit corresponding to the user profile as a second candidate unit; and at step S2030, a second list of recommended unit types is generated based on the second candidate unit types. Those skilled in the art will readily appreciate that, based on the user profile, the characteristics of the unit types that the user is interested in can be determined, and unit types with these characteristics can be searched accordingly. One or more unit types that meet the user's needs can then be selected as second candidate units. These one or more second candidate units are then listed in a list, thereby generating a second list of recommended unit types, which are ultimately recommended to the user.

[0047] Later, in Figure 3At step S2040 shown, it is determined whether the user has selected any apartment type in the second recommended apartment type list: if so, the method proceeds to S2050 to output the house source corresponding to the apartment type selected by the user. It is easy to understand that in the stored house sources, each house source can be divided into a certain apartment type category, that is, each apartment type has corresponding several houses. When the user selects a certain apartment type, the house source under the selected apartment type is displayed to the user; if not, the method proceeds to S2060 to determine whether the apartment type edited by the user has changed, that is, the user’s failure to select the recommended apartment type may mean that there is no ideal apartment type for the user. The user will also express his needs by drawing an apartment type. The change of the apartment type edited by the user means that the user has started to draw the apartment type or modified the drawn apartment type. In a preferred embodiment of the present invention, the user edits Changes in the edited apartment type include one or more of the following situations: addition, reduction, movement, rotation, and scaling of apartment type elements. Any of the above situations can indicate that the user has further drawn the apartment type: If so, the method proceeds to S2100 to extract the current apartment type, that is, the apartment type further drawn by the user is used as the current apartment type. Regardless of whether the current apartment type has been completely drawn as the final apartment type, the current apartment type can be extracted to monitor the user's apartment type needs in real time; if not, the method repeats the step of determining whether the apartment type edited by the user has changed in S2060, that is, if the user has neither selected any recommended apartment type nor edited the apartment type, it is repeatedly determined whether the apartment type edited by the user has changed until the user edits the apartment type again, and the changed apartment type is used as the current apartment type in step S2070. After step S2070, the same steps as above can be executed. Figure 2 The same steps S1100 to S1400 as shown in FIG, that is, recommending an apartment type to the user based on the current apartment type. In other words, Figure 3 The illustrated embodiment shows that after a user enters the intelligent apartment recommendation function, the user is recommended possible apartment types based on the acquired user profile. At this point, no features for drawing the apartment type are extracted, so the features here only refer to the user profile. If the user does not select a recommended apartment type, the user can start drawing the apartment type from scratch. During the drawing process, the server 105 will detect and extract the layout features of the current apartment type in real time, and then perform the most similar cluster search in the background (this will be explained in detail below).

[0048] exist Figure 3 In the embodiment shown, compared with Figure 2 The method shown also includes an additional step: determining whether the apartment type edited by the user has changed, that is, detecting whether the current apartment type has been modified: if so, the changed apartment type is used as the current apartment type; if not, repeatedly determining whether the apartment type edited by the user has changed. Specifically, Figure 1The current apartment type inputted by the user in step S1100 can be triggered based on detecting a change in the apartment type edited by the user. That is, when a change in the apartment type edited by the user is detected, the changed apartment type is extracted and then executed. Figure 1 If no changes are detected in the apartment type edited by the user, the detection is repeated continuously, that is, the apartment type edited by the user is continuously judged to be changed, until a change is detected. It should be noted that the step of judging whether the apartment type edited by the user has changed can be performed in Figure 2 , rather than just before step S1100, and Figure 3 At step S1050, when the user selects the recommended apartment type and the corresponding housing listing is output to the user, it can also be determined whether the apartment type edited by the user has changed. Because the monitoring of whether the apartment type edited by the user has changed is real-time and continuous, the latest round of apartment type recommendations will be triggered whenever a change is detected.

[0049] In the embodiment described above, the apartment editing tool can be configured to automatically detect changes in the apartment type edited by the user and recommend an apartment type to the user in real time. Alternatively, it can be configured to automatically detect changes in the apartment type edited by the user and recommend an apartment type to the user in real time based on the user logging into the apartment editing tool user interface or the user clicking Figure 4 The "Query" button in the lower right corner triggers the action of recommending apartment types to users.

[0050] Back to Figure 1 , method 1000 then obtains a first alternative apartment type similar to the current apartment type based on the current apartment type at step S1200. In a preferred embodiment of the present invention, this step also includes extracting the layout features of the current apartment type. In the real estate field, different apartment types have their own specific spatial layouts, such as the number of rooms, the type, location, area, shape, etc. of each room, which can constitute the layout features of the apartment type. Furthermore, in order to facilitate storage and use, the layout features of the apartment type can also be converted into vector values to obtain the characteristic vector of the apartment type. Those skilled in the art can vectorize the layout features using known methods, for example, by calculating the distance between any two compartments of the house and arranging the distances into an array, and then generating a vector for characterizing the compartment layout of the house based on the array. For the current apartment type, it can be an incomplete apartment type edited by the user using the user-type editing tool, but this does not affect the fact that it still has the various layout features mentioned above. It can still be used to retrieve apartment types similar to or containing the current apartment type based on the current layout features.

[0051] Subsequently, method 1000 can compare the layout characteristics of the current apartment type with the layout characteristics of apartment types in the apartment type library to find an apartment type similar to the current apartment type as the first candidate apartment type. Specifically, the apartment type library stores all types of apartment types, and each apartment type has its own layout characteristics. By comparing the layout characteristics of the current apartment type with the layout characteristics of apartment types in the apartment type library, or by other methods commonly used in the art, an apartment type similar to the layout characteristics of the current apartment type can be found in the apartment type library. For example, an apartment type in the apartment type library with a similarity greater than 80% can be selected as the first candidate apartment type.

[0052] Figure 5 Shown Figure 1 A preferred embodiment of step S1200 includes: at S1210, respectively quantizing the layout feature vectors of the housing types in the housing type library and the current housing type to obtain the feature vector of each housing type in the housing type library and the feature vector of the current housing type; at S1220, using a preset clustering method to divide the housing types in the housing type library into multiple housing type clusters; at S1230, calculating the central feature vector of each housing type cluster; at S1240, calculating the distance between the current housing type feature vector and the central feature vector of each housing type cluster to obtain the cluster distance; at S1250, extracting the housing type clusters whose cluster distance is less than a first threshold as similar housing type clusters; and at S1260, taking the housing type under the similar housing type cluster as the first alternative housing type.

[0053] Specifically, in Figure 5 At step S1210 shown, the layout features of the current apartment type can be vectorized as described above, and the layout features of the apartments in the apartment type library can also be vectorized in the same way, so that the feature vector of each apartment type in the apartment type library and the feature vector of the current apartment type can be obtained.

[0054] Later, in Figure 5At step S1220 shown, the preset clustering method is used to divide the apartment types in the apartment type library into multiple apartment type clusters: although each apartment type has different characteristics, the differences between some apartment types are large, while the differences between others are small. For example, the difference between two-bedroom apartment types is smaller than the difference between a two-bedroom apartment type and a three-bedroom apartment type. Based on this, all apartment types in the apartment type library can be divided into several major categories. In a preferred embodiment of the present invention, the mean-shift algorithm can be used to cluster the apartment types in the apartment type library to generate several apartment type clusters. The apartment types assigned to the same cluster need to have a layout distance less than a preset similarity threshold, and the similarity threshold can be set according to actual conditions and needs. In other words, the apartment types within the cluster need to be similar enough, while the apartment types outside the cluster need to be dissimilar enough. Alternatively, other methods or algorithms known to those skilled in the art can be used to cluster the apartment types in the apartment type library. Alternatively, this step can also be performed in advance instead of in the actual step. Figure 1 Executed in step S1200.

[0055] Later, in Figure 5 At step S1230, as those skilled in the art will readily understand, the apartment type drawn by the user may be one they have never seen before. Only the feature vector of the user's apartment type can retrieve the corresponding apartment type. Directly creating a vector index for all tens of millions of existing apartment types and returning it to the user specifically identifying a specific apartment type would result in a very long search response time. Moreover, such a search would be too specific and not very recommendable. In other words, the user may have only drawn a rough layout, leaving some details uncertain. Only clustering can summarize this uncertainty into a generalizable cluster. Therefore, the center vector of a cluster is used as the index of that cluster. Specifically, the apartment type at the center of each apartment type cluster can be used as the most representative apartment type in that cluster. Therefore, the feature vector of the apartment type at that center can be used as the central feature vector of the cluster. For example, for apartment type clusters obtained using the mean-shift algorithm, the apartment type at the highest density point in each cluster can be used as the representative apartment type, and the feature vector of this representative apartment type can be used as the central feature vector of the cluster. Alternatively, this step may be performed in advance like step S1220, instead of Figure 1 Executed in step S1200.

[0056] Later, in Figure 5At steps S1240 and S1250 shown, since the current apartment type feature vector and the central feature vector of each apartment type cluster are both vectors, those skilled in the art can use known methods to calculate the distance between the two, i.e., the cluster distance, which can represent the degree of similarity between the current apartment type and each cluster. Then, apartment type clusters whose cluster distance is less than a first threshold are extracted as similar apartment type clusters. The first threshold can be set according to actual conditions and needs. The apartment type clusters extracted here whose cluster distance is less than the first threshold are apartment type clusters with a greater degree of similarity to the current apartment type. The first threshold can be set according to actual conditions and needs, and this specification does not specifically limit this.

[0057] In another embodiment of the present invention, a spatial retrieval method can also be used to find housing type clusters similar to the current housing type. For example, based on vector space indexing methods such as Ball-tree and KD-tree, the central feature vectors of one or more housing type clusters that are nearest / similar to the current housing type feature vector can be retrieved, and then the housing type clusters where the one or more central feature vectors are located can be used as similar housing type clusters. This embodiment can avoid the heavy data processing process brought about by traditional pairwise comparison.

[0058] Later, in Figure 5 At step S1260, the unit type in the similar unit type cluster is selected as the first candidate unit type. In a preferred embodiment of the present invention, after the unit type clustering is completed, each unit type can be assigned a meaningless class number. Then, the unit types in the cluster can be used to establish a spatial index ID of each unit type to the central feature vector of the cluster, for example, through a Hashmap. In this way, once a cluster is determined to be a similar unit type cluster, each unit type in the cluster can be quickly indexed with an extremely high time complexity of O(1).

[0059] Subsequently, method 1000 generates a first recommended apartment type list based on the first alternative apartment type at step S1300, wherein an appropriate filtering method can be set to filter the apartment types in the first alternative apartment type for recommendation to the user, that is, the first recommended apartment type list can include all or part of the apartment types in the first alternative apartment type.

[0060] Figure 6 Shown Figure 1A preferred embodiment of step S1300 includes: obtaining a user portrait of the user at S1310, and vectorizing the user portrait to obtain a user portrait feature vector; calculating the distance between the feature vector of each apartment type in the first alternative apartment type and the feature vector of the current apartment type and the user portrait feature vector at S1320 to obtain an apartment type similarity distance and a user portrait distance; setting weights for the apartment type similarity distance and the user portrait distance at S1330, and calculating the weighted values of the two to obtain a total similarity distance; and extracting apartment types whose total similarity distance is less than a second threshold at S1340 as a first recommended apartment type list.

[0061] Specifically, in Figure 6 At step S1310 shown, the obtained user portrait can also be vectorized according to methods known to those skilled in the art to obtain a user portrait feature vector.

[0062] Later, in Figure 6 At step S1320 shown, since the difference between the first alternative apartment type - that is, the apartment type in the similar apartment type cluster - and the current apartment type is also different, the distance between the feature vector of each apartment type in the similar apartment type cluster and the feature vector of the current apartment type is calculated, and the similarity between each apartment type in the similar apartment type cluster and the current apartment type, that is, the apartment type similarity distance, can be obtained; and the distance between the feature vector of each apartment type in the similar apartment type cluster and the feature vector of the user portrait is calculated, and the similarity between each apartment type in the similar apartment type cluster and the user's ideal apartment type reflected by the user portrait, that is, the user portrait distance, can be obtained.

[0063] Later, in Figure 6 In step S1330, weights are set for the apartment type similarity distance and the user profile distance, and the weighted values of the two are calculated to obtain the total similarity distance. For example, the total similarity distance can be calculated using the following formula (1):

[0064] Distance=W1*Distance 1+W2*Distance 2 (1)

[0065] In the formula, Distance 1 represents the apartment type similarity distance, Distance 2 represents the user profile distance, W1 represents the weight of the apartment type similarity distance, and W2 represents the weight of the user profile distance. In the above embodiment, the specific values of W1 and W2 can be set according to actual conditions and needs, and the present invention does not specifically limit this. For example, the method of the present invention retrieves one or more apartment type clusters that are the nearest neighbors of the current apartment type through a spatial vector indexing method (such as Ball-tree, KD-tree, etc.), and the one or more apartment type clusters are respectively retrieved through a hashmap to bring out all the associated apartment types under these apartment type clusters. Then, the apartment type similarity distances between these associated apartment types and the current apartment type are calculated respectively. At the same time, the method of the present invention also predicts one or more apartment types that the user may want to find based on the user profile. Then, multiple apartment type similarity distances Distance 1 between all the associated apartment types under the retrieved one or more apartment type clusters and the current apartment type, as well as one or more user profile distances Distance 2 between the one or more apartment types predicted based on the user profile and the current apartment type are calculated respectively, and then multiple total similarity distances are calculated according to formula (1).

[0066] Later, in Figure 6 At step S1340, the unit types whose total similarity distance is less than the second threshold are extracted as the first recommended unit type list. Specifically, the multiple total similarity distances calculated above can be sorted from smallest to largest, and the clustered unit types and / or user profile-predicted unit types whose total similarity distance is less than the second threshold can be recommended to the user. Alternatively, the top n clustered unit types and / or user profile-predicted unit types in the total similarity distance sorting can be recommended to the user. The second threshold can be set based on actual conditions and needs, and this specification does not impose any specific restrictions on this.

[0067] For example, if the user draws two bedrooms on both sides of the living room and the user has a rigid demand for taking care of children, then by filtering all the cost-effective apartment types under the cluster, a list of apartment types closest to the user's intention will be recommended to the user through formula (1) for the user to choose. Or if the user still does not choose the recommended apartment type, but chooses to continue drawing the apartment type, then the execution can be repeated. Figure 3 Step S1060 determines whether the apartment layout edited by the user has changed. In other words, as long as the user draws or edits the apartment layout (the current apartment layout has changed), the method described in the present invention can extract the current apartment layout in real time to continuously understand the user's ideas and recommend corresponding apartment layouts.

[0068] Based on the same inventive concept, Figure 7The structure diagram of the intelligent apartment recommendation device 3000 according to an embodiment of the present invention is shown. The device 3000 includes: a receiving module 3100 configured to receive the current apartment type input by the user. In an embodiment of the present invention, the receiving module 3100 can be configured to execute Figure 2 S1100 shown in FIG. 1 and in this manual Figure 2 The step corresponding to S1100; the acquisition module 3200 is configured to obtain a first candidate apartment type similar to the current apartment type based on the current apartment type. In an embodiment of the present invention, the acquisition module 3200 can be configured to execute Figure 2 S1200, Figure 5 As shown and in this manual Figure 2 S1200, Figure 5 Corresponding steps; generating module 3300, which is configured to generate a first recommended apartment type list based on the first candidate apartment type. In an embodiment of the present invention, the generating module 3300 can be configured to execute Figure 2 S1300, Figure 6 As shown and in this manual Figure 2 S1300, Figure 6 The corresponding steps.

[0069] Based on the same inventive concept, Figure 8 The schematic diagram of the structure of the intelligent apartment recommendation device 4000 according to another embodiment of the present invention is shown. The device 4000 includes: an updating module 4400, which is configured to monitor whether the current apartment type edited by the user has changed: if so, the changed apartment type is used as the current apartment type; if not, the updated apartment type is continuously monitored for changes. In the embodiment of the present invention, the updating module 4400 can be configured to execute Figure 3 S2060, S2070 shown in FIG and in this manual Figure 3 Steps corresponding to S2060 and S2070; a receiving module 4100 configured to receive the current apartment type input by the user. In an embodiment of the present invention, the receiving module 4100 may be configured to execute Figure 2 S1100 shown in FIG. 1 and in this manual Figure 2 The step corresponding to S1100; the acquisition module 4200 is configured to obtain a first candidate apartment type similar to the current apartment type based on the current apartment type. In an embodiment of the present invention, the acquisition module 4200 can be configured to execute Figure 2 S1200, Figure 5 As shown and in this manual Figure 2 S1200, Figure 5Corresponding steps; generating module 4300, which is configured to generate a first recommended apartment type list based on the first candidate apartment type. In an embodiment of the present invention, the generating module 4300 can be configured to execute Figure 2 S1300, Figure 6 As shown and in this manual Figure 2 S1300, Figure 6 The corresponding steps.

[0070] I understand. Figures 7-8 The structure shown is only for illustration, and the device may also include Figures 7-8 More or fewer modules or components shown in, or with Figures 7-8 Different configurations shown.

[0071] In addition, the present application also provides a computer device. According to one embodiment of the present invention, the computer device may include a memory, a processor, and a computer program stored in the memory and run on the processor. When the processor executes the program, the steps of the intelligent apartment recommendation method described in this specification can be implemented.

[0072] In addition, this application also provides a computer-readable medium, which may be included in the apparatus described in the above embodiments, or may exist independently and not be incorporated into the apparatus. The computer-readable medium carries one or more programs, which, when executed by the apparatus, enable the apparatus to perform the steps of the intelligent apartment type recommendation method described in this specification.

[0073] In addition, the present application also provides a computer program product, including computer instructions, which, when executed by a processor, can implement the steps of the intelligent apartment type recommendation method described in this specification.

[0074] In particular, the embodiment processes described above with reference to the flowcharts in the accompanying drawings can be implemented as computer software programs. For example, the embodiments disclosed in this specification include a computer program product, which includes a computer program carried on a computer-readable medium, the computer program containing program code for executing the methods shown in the flowcharts in the accompanying drawings, and the computer program is executed by a processor to perform the methods of the present application.

[0075] It should be noted that the computer-readable medium described in this application may be a computer-readable signal medium or a computer-readable storage medium or any combination thereof. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: a computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0076] In this application, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. Furthermore, in this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transfer a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination thereof.

[0077] The computer program code for performing the operations of the present application can be written in one or more programming languages or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0078] The flow charts and block diagrams in the accompanying drawings exemplarily illustrate the possible architecture, functions and operations of the system, method and computer program product according to each embodiment of the present application. Each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code includes one or more executable instructions for realizing the specified logical function. It should be noted that in some alternative embodiments, the functions marked in the box can also occur in a different order than shown in the accompanying drawings. For example, two boxes represented in sequence can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a special hardware-based system that performs the specified function or operation, or can be implemented by a combination of special hardware and computer instructions.

[0079] The units or modules involved in the embodiments of the present application may be implemented in software or hardware. The above-mentioned units or modules may also be provided in a processor. For example, they may be described as: a processor including a receiving module, an acquisition module, a generation module, etc. The names of these units or modules do not, in some cases, constitute limitations on the units or modules themselves.

[0080] All documents mentioned in this specification are incorporated herein by reference, as if each document were incorporated herein by reference in its entirety.

[0081] In addition, it should be understood that after reading the above description of the present invention, those skilled in the art may make various changes or modifications to the present invention, and these equivalent forms also fall within the scope of protection of the present invention.

Claims

1. A method for intelligently recommending apartment types, characterized in that: include: A receiving step of receiving a current apartment type input by a user, wherein the current apartment type is used to express the user's demand for an apartment type; an acquisition step of acquiring a first candidate apartment type similar to the current apartment type based on the current apartment type; and A generating step of generating a first recommended apartment type list based on the first candidate apartment types; The current apartment type is edited by the user using a preset apartment type editing tool; The intelligent apartment type recommendation method further includes: In response to detecting a modification to the current apartment type, the obtaining step and the generating step are performed again on the modified current apartment type.

2. The method according to claim 1, characterized in that Also includes: Extract the layout features of the current apartment type, Acquiring a first candidate apartment type similar to the current apartment type based on the current apartment type includes: comparing the layout characteristics of the current apartment type with the layout characteristics of apartment types in an apartment type library, and searching for an apartment type similar to the current apartment type as the first candidate apartment type.

3. The method according to claim 1, characterized in that The preset apartment type editing tool has a user operation interface, which includes a drawing area and an element area. The user edits the apartment type by drawing in the drawing area using apartment type elements in the element area. The apartment type elements include different types of rooms.

4. The method according to claim 1, wherein The modification of the current apartment layout includes one or more of the following: adding, reducing, moving, rotating, and scaling apartment layout elements.

5. The method according to claim 2, characterized in that The apartment type library includes a plurality of apartment type clusters, each of which has a central feature vector; Comparing the layout characteristics of the current apartment type with the layout characteristics of apartment types in an apartment type library, searching for an apartment type similar to the current apartment type as a first candidate apartment type includes: Quantize the layout feature vectors of the apartment types in the apartment type library and the current apartment type respectively to obtain the feature vector of each apartment type in the apartment type library and the feature vector of the current apartment type; Calculating the distance between the current apartment type feature vector and the central feature vector of each apartment type cluster to obtain the cluster distance; Extracting apartment type clusters whose cluster distance is less than a first threshold as similar apartment type clusters; and The apartment type under the similar apartment type cluster is used as the first candidate apartment type.

6. The method according to claim 5, characterized in that Generating a first recommended apartment type list based on the first candidate apartment types includes: Obtaining a user portrait of the user, and vectorizing the user portrait to obtain a user portrait feature vector; Calculate the distance between the feature vector of each apartment type in the first candidate apartment type, the feature vector of the current apartment type, and the feature vector of the user portrait, respectively, to obtain an apartment type similarity distance and a user portrait distance; Set weights for the apartment similarity distance and user profile distance respectively, and calculate the weighted value of the two to obtain the total similarity distance; and The housing types whose total similarity distance is less than the second threshold are extracted as a first recommended housing type list.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: When the processor executes the program, the steps of the method according to any one of claims 1 to 6 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer program product comprising computer instructions, characterized in that When the computer instructions are executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images