Method, apparatus, computer device and storage medium recommended by real estate agent

Abstract

The present application discloses a method, device, computer equipment and storage medium for recommending a real estate agent. The method includes: receiving a viewing request sent by a target user, wherein the viewing request includes an expected viewing time and an expected viewing house; matching the viewing request with the viewing information of the real estate agent to obtain a candidate real estate agent; obtaining an attitude evaluation score of the candidate real estate agent; and determining a target real estate agent from the candidate real estate agents according to the attitude evaluation score. By implementing the embodiments of the present application, the accuracy of recommending real estate agents can be improved to help users find suitable real estate agents, thereby improving the viewing transaction rate and user experience.

Description

Technical Field

[0001] The present application relates to the field of information processing technology, and in particular to a method, device, computer equipment and storage medium for real estate agent recommendation. Background Art

[0002] In the real estate industry, real estate agents (intermediaries) play an indispensable role. When users have a need to buy a house, they usually need to contact real estate agents to arrange to view the house. After receiving the user's viewing request, the real estate platform often randomly assigns a real estate agent with time to show the user the house of interest. However, due to the uneven viewing ability of real estate agents, this random recommendation method leads to a low viewing transaction rate and poor user experience. Summary of the invention

[0003] The embodiments of the present application provide a method, apparatus, computer device, and storage medium for recommending real estate agents, which can improve the accuracy of recommending real estate agents to help users find suitable real estate agents, thereby improving the viewing transaction rate and user experience.

[0004] In a first aspect, an embodiment of the present application provides a method for real estate agent recommendation, comprising:

[0005] Receive a viewing request sent by a target user, wherein the viewing request includes a desired viewing time and a desired viewing property;

[0006] Matching the viewing request with the viewing information of the real estate agent to obtain candidate real estate agents;

[0007] Obtaining the attitude evaluation score of the candidate real estate agent;

[0008] A target real estate agent is determined from among the candidate real estate agents according to the attitude evaluation scores.

[0009] In a second aspect, an embodiment of the present application provides a device for real estate agent recommendation, including:

[0010] A communication unit, configured to receive a viewing request sent by a target user, wherein the viewing request includes an expected viewing time and an expected viewing house;

[0011] The processing unit is used to match the showing request with the showing information of the real estate agent to obtain candidate real estate agents; obtain the attitude evaluation score of the candidate real estate agents; and determine the target real estate agent from the candidate real estate agents according to the attitude evaluation score.

[0012] In a third aspect, an embodiment of the present application provides a computer device, comprising a processor, a memory, and a communication interface, wherein the memory stores a computer program, the computer program is configured to be executed by the processor, and the computer program includes instructions for some or all of the steps described in the first aspect of the embodiment of the present application.

[0013] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program enables a computer to execute some or all of the steps described in the first aspect of the embodiment of the present application.

[0014] Implementing the embodiments of the present application will have the following beneficial effects:

[0015] By using the above-mentioned method, device, computer equipment and storage medium for recommending real estate agents, after receiving a viewing request sent by a target user, the viewing request is matched with the viewing information of the real estate agent to determine a candidate real estate agent. Then, the attitude evaluation score of the candidate real estate agent is obtained, and the target real estate agent is determined from the candidate real estate agents according to the attitude evaluation score. In this way, by recommending a real estate agent based on the viewing request of the target user and the attitude evaluation score of the candidate real estate agent, the accuracy of recommending real estate agents can be improved, so as to help users find suitable real estate agents, thereby improving the viewing transaction rate and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0017] Figure 1 A schematic diagram of a system architecture provided for an embodiment of the present application;

[0018] Figure 2 A flowchart of a method for real estate agent recommendation provided in an embodiment of the present application;

[0019] Figure 3 A schematic diagram of the structure of a device recommended by a real estate agent provided in an embodiment of the present application;

[0020] Figure 4 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0022] The terms "first", "second", "third" and "fourth" etc. in the specification and claims of the present application and the drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0023] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0024] It should also be understood that the term "and / or" in this article is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0025] In order to better understand the technical solution of the embodiment of the present application, the system architecture that may be involved in the embodiment of the present application is first introduced. Figure 1 , a schematic diagram of a system architecture provided in an embodiment of the present application, the system architecture may include: an electronic device 101 and a server 102. The electronic device 101 and the server 102 may communicate via a network. The network communication may be based on any wired and wireless network, including but not limited to the Internet, a wide area network, a metropolitan area network, a local area network, a virtual private network (VPN), and a wireless communication network, etc.

[0026] The embodiment of the present application does not limit the number of electronic devices and servers, and the server can provide services for multiple electronic devices at the same time. In the embodiment of the present application, the electronic device can be used to receive the viewing request sent by the target user, and can also match the viewing request with the viewing information of the real estate agent to obtain a candidate real estate agent, and can also determine the target real estate agent among the candidate real estate agents according to the attitude evaluation score, etc. The electronic device can be a personal computer (personalcomputer, PC), a laptop or a smart phone, and can also be an all-in-one machine, a handheld computer, a tablet computer (pad), a smart TV playback terminal, a car terminal or a portable device, etc. The electronic device on the PC side, such as an all-in-one machine, etc., its operating system can include but is not limited to Linux system, Unix system, Windows series system (such as Windows xp, Windows 7, etc.), Mac OS X system (operating system of Apple computer) and other operating systems. The electronic device on the mobile side, such as a smart phone, etc., its operating system can include but is not limited to Android system, IOS (operating system of Apple mobile phone), Window system and other operating systems.

[0027] The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms. The server can also be implemented as a server cluster consisting of multiple servers.

[0028] In the real estate industry, real estate agents (intermediaries) play an indispensable role. When users have a need to buy a house, they usually need to contact real estate agents to arrange to view the house. After receiving the user's viewing request, the real estate platform often randomly assigns a real estate agent with time to show the user the house of interest. However, due to the uneven viewing ability of real estate agents, this random recommendation method leads to a low transaction rate and a poor user experience.

[0029] In order to solve the above problems, the embodiment of the present application provides a method for real estate agent recommendation, which can be applied to electronic devices or servers configured in the real estate field. By implementing this method, the accuracy of real estate agent recommendations can be improved, the user experience can be improved, and thus the viewing transaction rate can be increased.

[0030] Please refer to Figure 2 , Figure 2This is a flow chart of a real estate agent recommendation method provided in an embodiment of the present application. Taking the method applied to an electronic device as an example, the method may include the following steps S201-S204, wherein:

[0031] Step S201: receiving a viewing request sent by a target user.

[0032] In the embodiment of the present application, the target user may refer to a user who has a demand for buying a house and needs a real estate agent to show the house to him. The showing request may include the expected showing time and the expected showing source.

[0033] The expected viewing time may refer to the time when the target user wants the real estate agent to lead the tour of the house. The expected viewing time may be a certain time period of the day, or a certain time period a few days later. The expected viewing house may refer to the house that the target user wants the real estate agent to lead the tour, which may be specific property information, such as ABC New Town, etc.; or it may be information without specific property information, but with property preference requirements such as area, price, area, room type, orientation, floor, building age, decoration, elevator, surrounding public buildings, etc. For example, the viewing request may be "I hope to show the new house in ABC New Town from 10 am to 12 pm tomorrow", or it may be "I hope to show the second-hand house with three bedrooms and one living room from 2 pm to 4 pm the day after tomorrow", etc.

[0034] In addition, the viewing request may also include the target user's preference information for the real estate agent, such as the real estate agent's gender, personality, years of experience, performance and other preference information, and may also include the user's gender, name, contact information and other basic information. For example, the viewing request may be "I hope an enthusiastic female real estate agent will show me the new house in ABC New Town from 10 am to 12 pm this Saturday, contact number 123********" and so on.

[0035] Step S202: Match the viewing request with the viewing information of the real estate agent to obtain candidate real estate agents.

[0036] In an embodiment of the present application, the real estate agent's viewing information can be pre-stored in a preset real estate agent library, which can be pre-stored in an electronic device, or stored in a server, and the electronic device obtains the preset real estate agent library by accessing the server. The real estate agent's viewing information can include an estimated viewing time and an estimated viewing area. The estimated viewing time can refer to the time period when the real estate agent can lead the user to visit the house, and the estimated viewing area refers to the area where the real estate agent can lead the user to visit. The real estate agent's viewing information can also include other business information of the real estate agent, such as name, gender, age, personality traits, years of experience, historical viewing number, historical transaction rate, historical transaction total price, and contact information, etc.

[0037] In a possible implementation, step S202 may specifically include the following steps: searching for viewing information of all real estate agents from a preset real estate agent database, wherein the viewing information includes an expected viewing time and an expected viewing area; determining a real estate agent whose expected viewing time includes the expected viewing time and whose expected viewing area includes the location of the expected viewing house as a candidate real estate agent.

[0038] In an embodiment of the present application, the pre-stored real estate agent's showing information can be extracted from a preset real estate agent library, and the showing information can include an expected showing time and an expected showing area. Then, the real estate agent whose expected showing time includes the expected showing time and whose expected showing area includes the location where the expected showing house is located is searched. The real estate agent whose expected showing time includes the expected showing time and whose expected showing area includes the location where the expected showing house is located is selected as a candidate real estate agent. For example, if the expected showing time set by the user is from 10 am to 12 noon tomorrow, then the expected showing time of the real estate agent at least includes from 10 am to 12 noon tomorrow, or the time range is greater than from 10 am to 12 noon tomorrow, such as from 9 am to 4 pm tomorrow, etc. Similarly, the agent's expected showing area should also include the location where the expected showing house is located set by the user.

[0039] It can be seen that matching the viewing request with the real estate agent's viewing information, for example, determining the real estate agent whose estimated viewing time in the real estate agent's viewing information includes the desired viewing time and whose estimated viewing area includes the location of the desired viewing property, as a candidate real estate agent, can improve the matching efficiency of the candidate real estate agents, help users find suitable real estate agents, and improve the efficiency of property viewing.

[0040] In a possible embodiment, if the user does not explicitly select the desired viewing time or the desired viewing house in the viewing request, the real estate agent with the highest attitude evaluation score or viewing evaluation score among the real estate agents who are available in the current time period can be recommended to the target user. The specific calculation method of the attitude evaluation score and the viewing evaluation score of the real estate agent can refer to the description below and will not be repeated here.

[0041] In a possible embodiment, the expected viewing time in the viewing request can be extracted, and then the corresponding viewing requests can be sorted according to the order of the expected viewing time set by the user, that is, the viewing request corresponding to the expected viewing time closer to the current time is ranked higher, and the viewing request ranked higher is preferentially matched to the real estate agent. In this way, the inconvenience caused to the user by missing the user's expected viewing time can be avoided, which helps to improve the user experience.

[0042] Step S203: Obtain the attitude evaluation score of the candidate real estate agent.

[0043] The attitude evaluation score can be used to evaluate the service attitude of the candidate real estate agent. The more positive the service attitude of the candidate real estate agent is, the higher the attitude evaluation score is.

[0044] In a possible implementation, the attitude evaluation score can be obtained by obtaining user evaluation information. In an embodiment of the present application, after the user has finished showing or the transaction has ended, the user can push the attitude evaluation of the real estate agent who has served to the user, thereby obtaining the user evaluation information, and determining the attitude evaluation score according to the user evaluation information. For example, the service attitude can be divided into several different star ratings, and the attitude evaluation score is determined according to the star rating selected by the user. For example, the service level can be divided into five levels of five stars, four stars, three stars, two stars and one star, and the more stars, the more positive the service attitude. Then, different scores can be set in advance for each service attitude level, for example, the score area corresponding to five stars is 80-100, and the score areas corresponding to four stars, three stars, two stars and one star are 60-80, 40-60, 20-40 and 0-20 respectively. Finally, the attitude evaluation score of the candidate real estate agent is determined according to the score of the service attitude. In this way, the attitude evaluation score of the candidate real estate agent can be quickly obtained.

[0045] In addition, user evaluation information may also include user evaluation content, which can assist in evaluating the user's service attitude through the user evaluation content, thereby improving the accuracy of the attitude evaluation score. For example, if the comprehensive star rating of real estate agent A is four stars, then the attitude evaluation score is between 60 and 80 points. If positive keywords such as "enthusiasm" and "responsibility" are extracted from the user evaluation content of real estate agent A, the final attitude evaluation score of real estate agent A is determined to be in the high score area of ​​the score range of 60-80; if negative keywords such as "impolite" and "cold" are extracted from the user evaluation content of real estate agent A, the final attitude evaluation score of real estate agent A is determined to be in the low score area of ​​the score range of 60-80. In this way, the accuracy of the attitude evaluation score can be improved.

[0046] However, in actual applications, some users sometimes forget or are too lazy to comment on the service attitude of real estate agents, resulting in some of the real estate agents' attitude services being the system's default comments. The system's default comments make it difficult to objectively evaluate the service attitude of real estate agents, and it may make it difficult for the attitude evaluation score obtained based on user evaluation information to represent the service attitude of candidate real estate agents. Therefore, it is considered to calculate the attitude evaluation score of candidate real estate agents by performing text sentiment analysis on the chat data of users associated with candidate real estate agents.

[0047] Text sentiment analysis is the process of analyzing, processing, summarizing and reasoning about subjective texts with emotional connotations. By analyzing the texts, people's opinions and emotions expressed in the texts are obtained. At present, text sentiment analysis mainly focuses on the analysis of positive and negative emotions, that is, it is used to analyze whether the emotions expressed in natural language are positive emotions (positive emotions) or negative emotions (negative emotions); or, it also includes neutral emotions.

[0048] In a possible implementation, step S203 may specifically include steps A1-A5:

[0049] Step A1: Acquire text chat data of users associated with the candidate real estate agent.

[0050] In the embodiment of the present application, the user associated with the candidate real estate agent may be a user who has been associated with the candidate real estate agent, for example, a user who consulted the candidate real estate agent about house purchase issues or a user who was shown a house by the candidate real estate agent, etc. The text chat data may be the text conversation content between the associated user and the candidate real estate agent in the instant messaging (IM) tool of the real estate platform. The instant messaging tool may refer to an application software that realizes online chat and communication through instant messaging technology. The text chat data may include text data of both the real estate agent and the user, or may only include text data of the user.

[0051] In some possible implementations, step A1 may specifically include the following steps:

[0052] Acquire voice chat data of a user associated with the candidate real estate agent; perform frame processing on the voice chat data to obtain a first voice frame; perform preprocessing on the first voice frame to obtain a target voice frame; perform voice recognition on the target voice frame to obtain text chat data.

[0053] In the embodiment of the present application, when the real estate agent and the user communicate about the house purchase problem using the instant messaging tool, in addition to text chat, they can also use voice chat. Therefore, the voice chat data can be the voice conversation content between the associated user and the candidate real estate agent in the instant messaging tool (IM) of the real estate platform. Similar to the text chat data, the voice chat data can include the voice data of both the real estate agent and the user, or it can only include the voice data of the user.

[0054] The first voice frame refers to the voice frame obtained after the voice chat data is framed, and the target voice frame refers to the voice frame obtained after the first voice frame is preprocessed. Frame is the smallest observation unit in voice data. In the embodiment of the present application, framing can be understood as the process of dividing the voice chat data according to the temporal nature of the voice data. In order to make each frame signal in the voice chat data remain stable in a sufficiently short time and to meet the vibration cycle is sufficient, the frame length can be taken as 20-50ms, for example, 20ms, 25ms, 30ms, etc. In addition, in order to avoid excessive changes in two adjacent frames, the overlapping segmentation method can usually be used for framing, so that the frames transition smoothly and maintain their continuity. The overlapping part between frames is called frame shift, and the ratio of frame shift to frame length is generally taken as 0-1 / 2. In this way, by framing the voice chat data, a relatively stable voice frame can be obtained, which is conducive to improving the accuracy of subsequent speech recognition.

[0055] In an embodiment of the present application, the preprocessing method may include one or more of dimensionality reduction and normalization. Specifically, singular value decomposition (SVD), factor analysis (FA), principal component analysis (PCA), independent component analysis (ICA) and other methods may be used for dimensionality reduction. Through dimensionality reduction, the most important features can be retained from the high-dimensional vector, and noise and unimportant features can be removed. The first speech frame may also be normalized. For example, the min-max normalization method may be used to uniformly map the values ​​of the speech frames to the interval [0, 1]. In this way, the first speech frame after normalization has a certain degree of comparability in terms of numerical value, which can improve the accuracy of subsequent speech recognition.

[0056] In addition, the first speech frame obtained after the framing process can be subjected to windowing and pre-emphasis processing to obtain a target speech frame with better quality. Windowing can be used to eliminate signal discontinuity that may be caused at both ends of each frame, and can transform a non-stationary speech signal into a short-term stationary signal. Commonly used window functions include square window, Hamming window and Hanning window. Pre-emphasis can enhance the high-frequency part, which is used to filter out the low frequency and make the high frequency more prominent to improve the signal-to-noise ratio. After obtaining the target speech frame, the target speech frame can be converted into text chat data by using technologies such as automatic speech recognition (ASR).

[0057] It can be seen that after the voice chat data is framed, the first voice frame after the frame processing is preprocessed to obtain a target voice frame. Finally, the target voice frame is subjected to voice recognition to obtain text chat data. In this way, the efficiency and quality of voice recognition can be improved.

[0058] Step A2: performing word segmentation processing on the text chat data to obtain word segmentation vocabulary.

[0059] In an embodiment of the present application, the word segmentation processing method can adopt a word segmentation method based on string matching, also known as a mechanical word segmentation method. Word segmentation methods based on string matching include forward maximum matching method, reverse maximum matching method, shortest path word segmentation method and bidirectional maximum matching method. Among them, the forward maximum matching method refers to segmenting the string in a segmented sentence from left to right; the reverse maximum matching method refers to segmenting the string in a segmented sentence from right to left; the shortest path word segmentation method refers to the string in a segmented sentence requiring the minimum number of words to be segmented; the bidirectional maximum matching method refers to simultaneous forward and reverse word segmentation matching. The word meaning segmentation method can also be used to segment each segmented sentence. The word meaning segmentation method is a word segmentation method for machine speech judgment, which uses syntactic information and semantic information to handle ambiguous phenomena to segment words.

[0060] Step A3: extracting words associated with emotions from the segmented words as keywords.

[0061] Among them, the vocabulary associated with emotion can refer to the vocabulary that can reflect emotions from a certain height. In a possible implementation, an emotion dictionary library associated with emotion can be pre-established, in which the words that can reflect emotions, such as "happy", "glad", "like", "passion", "careful", "dedicated", "polite" and other words are included to indicate that the emotion of the text is negative. "Rude", "bad", "bad", "hate", "cold", "sad", "angry", "ridiculous" and other words indicate that the emotion of the text is negative and negative. After obtaining the segmentation vocabulary, the segmentation vocabulary obtained can be compared with the words included in the emotion dictionary library in turn to determine the segmentation vocabulary associated with emotion as a keyword. For example, "the brother who showed the house today is very attentive" can compare the segmentation vocabulary such as "today", "showing", "brother", "very", "careful" with the words included in the emotion dictionary library in turn, wherein "careful" is consistent with the words included in the emotion dictionary library, then "careful" can be used as a keyword. Keywords can be extracted from user chat content, or from the chat content of both the real estate agent and the user, and the embodiment of the present application does not limit this.

[0062] Step A4: Perform emotion recognition on the keywords to obtain text emotions.

[0063] Text emotions may include positive emotions (positive emotions), negative emotions (negative emotions), and neutral emotions. In an embodiment of the present application, emotion recognition is performed on each extracted keyword to determine whether the text emotion of the keyword is positive emotion or negative emotion. In a possible implementation, a preset text emotion recognition model can be used to identify the text emotion of the keyword. The text emotion recognition model may include one or more of a long short-term memory network (LSTM), a gated recurrent unit network (GRU), or an extreme gradient boosting (Xgboost) model, and the embodiment of the present application does not limit this.

[0064] In a possible implementation, there may be some negative words in the text chat data, and negative words may cause the text sentiment of the keyword to show opposite emotional characteristics. In specific applications, the negative words can be searched to see if they are included in the word segmentation vocabulary. If negative words are identified, the corresponding text sentiment of the keyword is adjusted in the opposite direction. For example, the target user's answer to the question is "The guy who showed me the place today was very rude." In this sentence, the keyword is "politeness", but the negative word "no" is retrieved. If the text sentiment of "politeness" is identified as a positive emotion, the text sentiment of "politeness" is adjusted to a negative emotion.

[0065] Step A5: Calculate the attitude evaluation score of the candidate real estate agent according to the text sentiment.

[0066] In an embodiment of the present application, the number of keywords can be counted first, and then the number of keywords with positive text emotions can be counted to calculate the proportion of positive emotions. The proportion of positive emotions is the ratio of the number of keywords with positive text emotions to the number of keywords. For example, the number of keywords included in the text chat data of the user associated with candidate real estate agent A is 100, and the number of keywords with positive text emotions is 95, then it can be calculated that the proportion of positive emotions is 95%. Finally, the attitude evaluation score of the candidate real estate agent is determined according to the proportion of positive emotions.

[0067] In a possible implementation, the proportion of positive emotions may be divided into different intervals in advance, and different intervals may be set to correspond to different attitude evaluation scores, for example, a positive emotion proportion of 80-100% corresponds to an attitude evaluation score range of 80-100, a positive emotion proportion of 60-80% corresponds to an attitude evaluation score range of 60-80, a positive emotion proportion of 0-20% corresponds to an attitude evaluation score range of 0-20, etc. That is, assuming that the proportion of positive emotions of candidate real estate agent A is calculated to be 95%, the attitude evaluation score of candidate real estate agent A may be 95.

[0068] It can be seen that by obtaining the text chat data of the users associated with the candidate real estate agent, and then performing word segmentation processing on the text chat data, the word segmentation vocabulary is obtained, and the words associated with emotions are extracted from the word segmentation vocabulary as keywords. Then, emotion recognition is performed on the keywords to obtain text emotions, so that the emotion recognition can have good accuracy. Finally, the attitude evaluation score of the candidate real estate agent is calculated based on the text emotion. The accuracy of the attitude evaluation score of the candidate real estate agent calculated based on the text emotion is also high. In addition, determining the attitude evaluation of the candidate real estate agent based on the text emotion can increase the frequency of use of positive words in the process of communicating with customers, guide users to give positive responses, thereby increasing transaction volume, allowing users to buy houses faster, and improving user experience.

[0069] In a possible implementation manner, after step A4, the following steps may also be included:

[0070] Perform feature extraction on the target voice frame to obtain target voice features; perform emotion recognition on the voice chat data based on the target voice features to obtain voice emotions; determine a comprehensive emotion based on the voice emotions and the text emotions; and calculate an attitude evaluation score of the candidate real estate agent based on the comprehensive emotion.

[0071] In the embodiment of the present application, due to the existence of voice chat data, the emotions actually expressed by people in the voice and the emotions expressed in the corresponding text of the voice may be quite different. For example, in some sarcastic scenes, the result of voice emotion recognition may be negative emotions, while the corresponding text of the voice may be recognized as positive emotions because it contains positive words. Therefore, in order to improve the accuracy of emotion recognition, in addition to text emotions, the recognition of voice emotions can be added.

[0072] In the embodiment of the present application, the target speech features may include but are not limited to mel-frequency cepstral coefficients (MFCC), identity-vector (i-vector), pitch, tone intensity, tone length, frequency band energy distribution, harmonic signal-to-noise ratio, short-term energy jitter, etc. Among them, MFCC has good robustness, conforms to the auditory characteristics of the human ear, and can still have good recognition performance when the signal-to-noise ratio is reduced. For the convenience of description, the extraction process of the target speech features is introduced below using MFCC as an example.

[0073] In a possible implementation, feature extraction is performed on a target speech frame to obtain a target speech feature, which may specifically include the following steps: performing fast Fourier transform processing on the target speech frame to obtain speech spectrum data; inputting the speech spectrum data into a Mel filter to obtain Mel frequency data; performing cepstrum analysis processing on the Mel frequency data to obtain Mel frequency cepstrum coefficients; and using the Mel frequency cepstrum coefficients as the target speech feature.

[0074] In the embodiment of the present application, the target voice frame is a signal obtained after the voice chat data is processed by framing, and it is still a time domain signal at this time. However, it is difficult to see the characteristics of the signal in the time domain signal, so it is necessary to convert the time domain signal into a frequency domain signal. Fast Fourier transform (fast Fourier transform, FFT) is a general term for the fast calculation of discrete Fourier transform (discreteFourier transform, DFT). After FFT processing, the time domain signal of the target voice frame can be converted into a frequency domain signal of the voice spectrum data. The Mel filter can smooth the voice spectrum data and play a role in eliminating filtering, highlighting the resonance peak characteristics of the voice. Finally, the Mel spectrum data is subjected to cepstrum analysis to obtain MFCC as a voice feature, and the cepstrum analysis can be implemented by discrete cosine transform (discrete cosine transform, DCT). DCT is a transform related to Fourier transform, similar to DFT, but DCT only uses real numbers. DCT can be used to remove the correlation between signals of each dimension and map the signal to a low-dimensional space. In this way, the Mel frequency cepstrum coefficients can be obtained with good robustness, which is conducive to improving the accuracy of subsequent voice emotion recognition.

[0075] Similar to text emotions, speech emotions can also include positive emotions (positive emotions), negative emotions (negative emotions), and neutral emotions. In an embodiment of the present application, emotion recognition is performed on the target speech feature to determine whether the text emotion of the keyword is a positive emotion or a negative emotion. In a possible implementation, a preset speech emotion recognition model can be used to identify the speech emotion of the target speech feature. The speech emotion recognition model may include one or more of a long short-term memory network (LSTM), a gated recurrent unit network (GRU), or an extreme gradient boosting (Xgboost), a recurrent neural network (RNN), a convolutional neural network (CNN) and other models, and the embodiment of the present application does not limit this.

[0076] In an embodiment of the present application, a corresponding rule for determining a comprehensive emotion based on voice emotion and text emotion can be pre-set, and the comprehensive emotion is determined according to the corresponding rule. Specifically, when the chat data is voice chat data, if the voice emotion in the voice chat data and the text emotion of the text chat data corresponding to the voice chat data show the same emotional polarity, then the comprehensive emotion is the same as the voice emotion and the text emotion. If the voice emotion in the voice chat data and the text emotion of the text chat data corresponding to the voice chat data show opposite emotional polarities, then the comprehensive emotion is the same as the voice emotion. For example, if the voice emotion in the voice chat data is a negative emotion, and the text emotion of the text chat data corresponding to the voice chat data is a positive emotion, then the comprehensive emotion is a negative emotion. Or it can also be other corresponding rules, such as pre-setting the weight of the voice emotion in the voice chat data and the weight of the text emotion of the text chat data corresponding to the voice chat data, respectively, and obtaining the comprehensive emotion by weighted calculation according to the preset weights, which is not limited in the embodiment of the present application.

[0077] After obtaining the comprehensive emotion, the attitude evaluation score can be determined according to the comprehensive emotion. The proportion of positive emotions in the comprehensive emotion can be counted, and then the proportion of positive emotions can be divided into different intervals, and different intervals can be set to correspond to different attitude evaluation scores. This method is similar to step A5 in calculating the attitude evaluation score of the candidate real estate agent according to the text emotion. For details, please refer to the description of step A5, which will not be repeated here.

[0078] It can be seen that after obtaining the voice chat data of the user associated with the candidate real estate agent, the voice chat data is first converted into text chat data. After that, emotion recognition is performed on the voice chat data and the text chat data respectively to determine the voice emotion and the text emotion. Then, the comprehensive emotion is determined based on the voice emotion and the text emotion. Finally, the attitude evaluation score of the candidate real estate agent is calculated based on the comprehensive emotion. Since emotion recognition is performed based on the data of both voice chat data and text chat data, the final result of emotion recognition can represent the voice chat data and the text chat data. Therefore, the comprehensive emotion obtained based on the voice emotion and the text emotion has better accuracy, which helps to improve the accuracy of the calculation result when the comprehensive emotion is used to calculate the attitude evaluation score in the subsequent process.

[0079] Step S204: determining a target real estate agent from among the candidate real estate agents according to the attitude evaluation scores.

[0080] The target real estate agent may refer to the real estate agent who is ultimately recommended to the target user. In a possible implementation, step S204 may specifically include the following steps:

[0081] If the number of candidate real estate agents corresponding to the maximum value of the attitude evaluation score is 1, the candidate real estate agent corresponding to the maximum value of the attitude evaluation score is used as the target real estate agent; if the number of candidate real estate agents corresponding to the maximum value of the attitude evaluation score is greater than 1, the candidate real estate agent corresponding to the maximum value of the attitude evaluation score is used as a reference real estate agent; the viewing evaluation score of the reference real estate agent is obtained, and the viewing evaluation score is used to evaluate the viewing success rate of the reference real estate agent; the reference real estate agent corresponding to the maximum value of the viewing evaluation score is used as the target real estate agent.

[0082] In the embodiment of the present application, the calculated attitude evaluation scores of the candidate real estate agents are sorted in descending order to determine the candidate real estate agent corresponding to the maximum attitude evaluation score. If the number of candidate real estate agents corresponding to the maximum attitude evaluation score is 1, for example, candidate real estate agent A is the candidate real estate agent corresponding to the maximum attitude evaluation score, then candidate real estate agent A is used as the target real estate agent.

[0083] If the number of candidate real estate agents corresponding to the maximum value of the attitude evaluation score is greater than 1, that is to say, more than one candidate real estate agent has achieved the maximum value of the attitude evaluation score at the same time. In the embodiment of the present application, if the number of candidate real estate agents corresponding to the maximum value of the attitude evaluation score is more than one, the two or more candidate real estate agents corresponding to the maximum value of the attitude evaluation score are called reference real estate agents. For example, candidate real estate agent B and candidate real estate agent C are both candidate real estate agents corresponding to the maximum value of the attitude evaluation score, then the reference real estate agent is candidate real estate agent B and candidate real estate agent C. For the convenience of description, candidate real estate agent B and candidate real estate agent C can be recorded as reference real estate agent B and reference real estate agent C. At this time, it can be considered to add dimensions such as the viewing evaluation score, and the reference real estate agent corresponding to the maximum value of the viewing evaluation score is used as the target real estate agent. For example, the viewing evaluation score of reference real estate agent B is 90 points, the viewing evaluation score of reference real estate agent C is 85 points, and the viewing evaluation score of reference real estate agent B is greater than the viewing evaluation score of reference real estate agent C. Reference real estate agent B can be used as the target real estate agent as the real estate agent finally recommended to the target user.

[0084] It can be seen that when the number of candidate real estate agents corresponding to the maximum value of the attitude evaluation score is 1, the candidate real estate agent corresponding to the maximum value of the attitude evaluation score is used as the target real estate agent. In this way, the target real estate agent can be quickly determined. When the number of candidate real estate agents corresponding to the maximum value of the attitude evaluation score is greater than 1, the candidate real estate agent corresponding to the maximum value of the attitude evaluation score is used as a reference real estate agent, and dimensions such as the viewing evaluation score can be added to determine the target real estate agent from the reference real estate agents. For example, the reference real estate agent corresponding to the maximum value of the viewing evaluation score can be used as the target real estate agent as the real estate agent finally recommended to the target user. In this way, by adding the dimension of the viewing evaluation score, the agent recommendation algorithm is improved, the accuracy of the recommended real estate agent is improved, and users can find suitable real estate agents, thereby improving the viewing transaction rate.

[0085] The viewing evaluation score can be used to evaluate the viewing success rate of the reference real estate agent. In a possible implementation, obtaining the viewing evaluation score of the reference real estate agent may include the following steps:

[0086] Obtain the historical number of viewings and the historical number of transactions of the reference real estate agent; obtain the historical transaction rate of the reference real estate agent, wherein the historical transaction rate is calculated based on the historical number of viewings and the historical number of transactions; and calculate a viewing evaluation score based on the historical transaction rate.

[0087] In the embodiment of the present application, the number of historical viewings refers to the number of times the reference real estate agent has led the user to visit the house as of the current moment, and the number of historical transactions refers to the number of transactions reached between the reference real estate agent and the user as of the current moment. The historical transaction rate may refer to the probability of the reference real estate agent's viewing and transaction as of the current moment. The historical transaction rate is calculated based on the historical number of viewings and the historical number of transactions of the reference real estate agent. For example, if the historical number of viewings of reference real estate agent B is 100 and the historical number of transactions is 90, then the historical transaction rate of reference real estate agent B may be 90 / 100*100%=90%, and the historical number of viewings of reference real estate agent C is 100 and the historical number of transactions is 85, then the historical transaction rate of reference real estate agent C may be 85 / 100*100%=85%.

[0088] The viewing evaluation score can be determined based on the historical transaction rate. For example, the historical transaction rate can be divided into different intervals in advance, and different intervals can be set to correspond to different viewing evaluation scores. For example, the viewing evaluation score area corresponding to the historical transaction rate of 80-100% is 80-100, the viewing evaluation score area corresponding to the historical transaction rate of 60-80% is 60-80, the viewing evaluation score area corresponding to the historical transaction rate of 0-20% is 0-20, and so on. In other words, if the historical transaction rate of reference real estate agent B is calculated to be 90%, then the viewing evaluation score of reference real estate agent B can be 90.

[0089] It can be seen that the historical transaction rate is calculated based on the historical viewing number and historical transaction number of the reference real estate agent, and the viewing evaluation score is calculated based on the historical transaction rate. In this way, the viewing evaluation score can be quickly calculated, and the accuracy of the viewing evaluation score can be improved, so as to improve the accuracy of the recommended real estate agent in the future.

[0090] In a possible implementation, when calculating the viewing evaluation score, in addition to considering the viewing transaction rate, it is also possible to consider adding dimensions such as the historical total transaction price and the historical unpopular property transaction rate to determine the viewing evaluation score. For example, when the viewing transaction rate is the same, the more historical viewings there are, the higher the viewing evaluation score; or when the viewing transaction rate is the same, the higher the historical total transaction price, the higher the viewing evaluation score; or when the viewing transaction rate is the same, the higher the historical unpopular property transaction rate, the higher the viewing evaluation score. When the viewing transaction rate is the same, the historical viewing times of the same user among the transaction users can also be obtained. The fewer the historical viewing times, the higher the viewing evaluation score. In this way, the calculated viewing evaluation score can have a higher accuracy.

[0091] Alternatively, in a possible implementation, when calculating the viewing evaluation score, preset weights are set in advance for the viewing transaction rate, the historical viewing number, the historical transaction total price, the historical unpopular real estate, etc., and the target viewing evaluation rate is obtained by weighted calculation based on the preset weights and the viewing transaction rate, the historical viewing number, the historical transaction total price, and the historical unpopular real estate, and the viewing evaluation score is determined based on the target viewing rate. The preset weights of the viewing transaction rate, the historical viewing number, the historical transaction total price, and the historical unpopular real estate can be determined based on historical experience, or can also be determined based on actual conditions. For example, the preset weight of the viewing transaction rate can be set to 0.9, the preset weight of the historical viewing total price can be set to 0.85, the preset weight of the historical viewing number can be set to 0.8, and the preset weight of the historical unpopular real estate can be set to 0.75, and the target viewing evaluation rate can be obtained by weighted calculation. Finally, the viewing evaluation score is calculated based on the target viewing evaluation rate, wherein the higher the target viewing evaluation rate, the higher the viewing evaluation score. In this way, the accuracy of the viewing evaluation score can be further improved.

[0092] In a possible implementation manner, after step S202, the following steps may also be included:

[0093] Obtain attitude evaluation scores and viewing evaluation scores of the candidate real estate agents; and determine a target real estate agent from among the candidate real estate agents based on the attitude evaluation scores and the viewing evaluation scores.

[0094] In an embodiment of the present application, after determining the candidate real estate agent, the attitude evaluation score and the viewing evaluation score of the candidate real estate agent can be obtained at the same time. The specific implementation method of obtaining the attitude evaluation score and the viewing evaluation score of the candidate real estate agent can refer to the description in the previous text, which will not be repeated here. After obtaining the attitude evaluation score and the viewing evaluation score of the candidate real estate agent, a comprehensive evaluation score is obtained according to the attitude evaluation score and the viewing evaluation score. The comprehensive evaluation score can be obtained by simply adding and calculating the attitude evaluation score and the viewing evaluation score. Alternatively, the attitude evaluation score and the viewing evaluation score can be set with preset weights in advance, and the comprehensive evaluation score is obtained by weighted calculation according to the preset weights. The preset weights of the attitude evaluation score and the viewing evaluation score can be determined according to historical experience, or can also be determined according to actual conditions. For example, the preset weight of the attitude evaluation score can be set to 0.9, and the preset weight of the viewing evaluation score can be set to 0.8, and finally the comprehensive evaluation score is obtained by weighted calculation. In this way, the accuracy of recommending real estate agents can be further improved, thereby improving the viewing transaction rate and user experience.

[0095] exist Figure 2In the method shown, after receiving the viewing request sent by the target user, the viewing request is matched with the viewing information of the real estate agent to determine the candidate real estate agent. Then, the attitude evaluation score of the candidate real estate agent is obtained, and the target real estate agent is determined from the candidate real estate agents according to the attitude evaluation score. In this way, the real estate agent is recommended based on the viewing request of the target user and the attitude evaluation score of the candidate real estate agent, which can improve the accuracy of the recommended real estate agent, help users find suitable real estate agents, and thus improve the viewing transaction rate and user experience.

[0096] It can be understood that in the specific implementation of this application, data related to the user's chat content, etc. is involved. When the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0097] The method of the embodiment of the present application is described in detail above, and the device of the embodiment of the present application is provided below.

[0098] Please refer to Figure 3 , Figure 3 1 is a schematic diagram of a structure of a device recommended by a real estate agent provided in an embodiment of the present application. The device is applied to electronic equipment. Figure 3 As shown, the device 300 recommended by the real estate agent includes a communication unit 301 and a processing unit 302. The detailed description of each unit is as follows:

[0099] The communication unit 301 is used to receive a viewing request sent by a target user, wherein the viewing request includes a desired viewing time and a desired viewing house;

[0100] The processing unit 302 is used to match the viewing request with the viewing information of the real estate agent to obtain candidate real estate agents; obtain the attitude evaluation scores of the candidate real estate agents; and determine the target real estate agent from the candidate real estate agents according to the attitude evaluation scores.

[0101] In one possible implementation, the processing unit 302 is specifically used to obtain text chat data of users associated with the candidate real estate agent; perform word segmentation processing on the text chat data to obtain a segmentation vocabulary; extract words associated with emotions from the segmentation vocabulary as keywords; perform emotion recognition on the keywords to obtain text emotions; and calculate the attitude evaluation score of the candidate real estate agent based on the text emotions.

[0102] In a possible implementation, the processing unit 302 is specifically used to obtain voice chat data of a user associated with the candidate real estate agent; perform frame processing on the voice chat data to obtain a first voice frame; perform preprocessing on the first voice frame to obtain a target voice frame; perform voice recognition on the target voice frame to obtain text chat data.

[0103] In a possible implementation, the processing unit 302 is further used to perform feature extraction on the target voice frame to obtain target voice features; perform emotion recognition on the voice chat data based on the target voice features to obtain voice emotions; determine a comprehensive emotion based on the voice emotions and the text emotions; and calculate the attitude evaluation score of the candidate real estate agent based on the comprehensive emotion.

[0104] In a possible implementation, the processing unit 302 is specifically used to, if the number of candidate real estate agents corresponding to the maximum value of the attitude evaluation score is 1, then use the candidate real estate agent corresponding to the maximum value of the attitude evaluation score as the target real estate agent; if the number of candidate real estate agents corresponding to the maximum value of the attitude evaluation score is greater than 1, then use the candidate real estate agent corresponding to the maximum value of the attitude evaluation score as a reference real estate agent; obtain the viewing evaluation score of the reference real estate agent, the viewing evaluation score is used to evaluate the viewing success rate of the reference real estate agent; and use the reference real estate agent corresponding to the maximum value of the viewing evaluation score as the target real estate agent.

[0105] In a possible implementation, the processing unit 302 is specifically used to obtain the historical number of viewings and the historical number of transactions of the reference real estate agent; obtain the historical transaction rate of the reference real estate agent, the historical transaction rate being calculated based on the historical number of viewings and the historical number of transactions; and calculate a viewing evaluation score based on the historical transaction rate.

[0106] In a possible implementation, the processing unit 302 is specifically used to search for viewing information of all real estate agents from a preset real estate agent database, wherein the viewing information includes an expected viewing time and an expected viewing area; and a real estate agent whose expected viewing time includes the expected viewing time and whose expected viewing area includes the location of the expected viewing house is determined as a candidate real estate agent.

[0107] It should be noted that the implementation of each unit can also refer to Figure 2 The corresponding description of the method embodiment shown.

[0108] Please refer to Figure 4 , Figure 4 Schematic diagram of the structure of a computer device provided in an embodiment of the present application. Figure 4As shown, the computer device 400 includes a processor 401, a memory 402 and a communication interface 403, wherein the memory 402 stores a computer program 404. The processor 401, the memory 402, the communication interface 403 and the computer program 404 may be connected via a bus 405.

[0109] When the computer device is an electronic device, the computer program 404 is used to execute instructions of the following steps:

[0110] Receive a viewing request sent by a target user, wherein the viewing request includes a desired viewing time and a desired viewing property;

[0111] Matching the viewing request with the viewing information of the real estate agent to obtain candidate real estate agents;

[0112] Obtaining the attitude evaluation score of the candidate real estate agent;

[0113] A target real estate agent is determined from among the candidate real estate agents according to the attitude evaluation scores.

[0114] In a possible implementation, in terms of obtaining the attitude evaluation score of the candidate real estate agent, the computer program 404 is specifically used to execute the following steps:

[0115] Obtaining text chat data of users associated with the candidate real estate agent;

[0116] Performing word segmentation processing on the text chat data to obtain a word segmentation vocabulary;

[0117] Extracting words associated with emotions from the word segmentation vocabulary as keywords;

[0118] Performing emotion recognition on the keywords to obtain text emotions;

[0119] An attitude evaluation score of the candidate real estate agent is calculated based on the text sentiment.

[0120] In a possible implementation, in the aspect of obtaining the text chat data of the user associated with the candidate real estate agent, the computer program 404 is specifically used to execute the following steps:

[0121] Acquire voice chat data of a user associated with the candidate real estate agent;

[0122] Performing frame processing on the voice chat data to obtain a first voice frame;

[0123] Preprocessing the first speech frame to obtain a target speech frame;

[0124] Perform speech recognition on the target speech frame to obtain text chat data.

[0125] In a possible implementation, after performing emotion recognition on the keywords to obtain text emotions, the computer program 404 is further used to execute instructions for the following steps:

[0126] Extracting features from the target speech frame to obtain target speech features;

[0127] Performing emotion recognition on the voice chat data according to the target voice features to obtain voice emotions;

[0128] Determining a comprehensive emotion according to the voice emotion and the text emotion;

[0129] An attitude evaluation score of the candidate real estate agent is calculated based on the comprehensive sentiment.

[0130] In a possible implementation, in determining a target real estate agent from among the candidate real estate agents according to the attitude evaluation scores, the computer program 404 is specifically used to execute instructions for the following steps:

[0131] If the number of candidate real estate agents corresponding to the maximum value of the attitude evaluation score is 1, the candidate real estate agent corresponding to the maximum value of the attitude evaluation score is used as the target real estate agent;

[0132] If the number of candidate real estate agents corresponding to the maximum value of the attitude evaluation score is greater than 1, the candidate real estate agent corresponding to the maximum value of the attitude evaluation score is used as a reference real estate agent;

[0133] Obtaining a viewing evaluation score of the reference real estate agent, wherein the viewing evaluation score is used to evaluate the viewing success rate of the reference real estate agent;

[0134] The reference real estate agent corresponding to the maximum viewing evaluation score is used as the target real estate agent.

[0135] In a possible implementation, in the aspect of obtaining the viewing evaluation score of the reference real estate agent, the computer program 404 is specifically used to execute the following steps:

[0136] Obtain the historical viewing quantity and historical transaction quantity of the reference real estate agent;

[0137] Obtaining a historical transaction rate of the reference real estate agent, where the historical transaction rate is calculated based on the historical number of viewings and the historical number of transactions;

[0138] The viewing evaluation score is calculated based on the historical transaction rate.

[0139] In a possible implementation, in the aspect of matching the showing request with the showing information of the real estate agent to obtain the candidate real estate agent, the computer program 404 is specifically used to execute the following steps:

[0140] Searching for viewing information of all real estate agents from a preset real estate agent database, wherein the viewing information includes an estimated viewing time and an estimated viewing area;

[0141] The real estate agent whose estimated showing time includes the desired showing time and whose estimated showing area includes the location where the desired showing house is located is determined as a candidate real estate agent.

[0142] Those skilled in the art will appreciate that for ease of description, Figure 4 Only one memory and processor are shown. In an actual terminal or server, there may be multiple processors and memories. The memory 402 may also be referred to as a storage medium or a storage device, etc., which is not limited in the present embodiment of the application.

[0143] It should be understood that in the embodiment of the present application, the processor 401 can be a central processing unit (CPU), and the processor can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0144] It should also be understood that the memory 402 mentioned in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronize link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0145] It should be noted that when the processor 401 is a general-purpose processor, DSP, ASIC, FPGA or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, the memory (storage module) is integrated in the processor.

[0146] It should be noted that the memory 402 described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0147] The bus 405 may include, in addition to the data bus, a power bus, a control bus, a status signal bus, etc. However, for the sake of clarity, various buses are labeled as buses in the figure.

[0148] In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in a processor or an instruction in the form of software. The steps of the method disclosed in conjunction with the embodiment of the present application can be directly embodied as a hardware processor for execution, or a combination of hardware and software modules in a processor for execution. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in conjunction with its hardware. To avoid repetition, it is not described in detail here.

[0149] In various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0150] Those skilled in the art will appreciate that the various illustrative logical blocks (ILBs) and steps described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0151] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0152] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0153] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0154] In the above embodiments, it 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 instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integration. The available medium can be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk), etc.

[0155] In the above embodiment, the computer-readable storage medium may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function, etc.; the data storage area may store data created according to the use of the blockchain node, etc. For example, the blockchain may store a real estate agent library, and may also store automatic speech recognition technology (automatic speech recognition, ASR), long short-term memory network (long shorttermmemory, LSTM), gated recurrent unit network (gated recurrent unit, GRU), or extreme gradient boosting (extreme gradient boosting, Xgboost), recurrent neural network (recurrent neural network, RNN), convolutional neural network (convolutional neural network, CNN) algorithm, etc., without limitation here.

[0156] Among them, the blockchain referred to in the embodiments of the present application is a new application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods, each of which contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0157] An embodiment of the present application also provides a computer storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement part or all of the steps of any real estate agent recommendation method recorded in the above method embodiments.

[0158] An embodiment of the present application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute part or all of the steps of any real estate agent recommendation method recorded in the above method embodiments.

[0159] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for real estate agent recommendation, characterized in that: include: Receiving a viewing request sent by a target user, wherein the viewing request includes a desired viewing time, a desired house to be viewed, and desired preference information for a real estate agent; Matching the viewing request with the viewing information of the real estate agent to obtain candidate real estate agents; If it is determined that there is no user evaluation information, obtaining voice chat data of a user associated with the candidate real estate agent, wherein the user evaluation information includes a user rating level and a user rating content; Performing frame processing on the voice chat data to obtain a first voice frame; Preprocessing the first speech frame to obtain a target speech frame; Performing speech recognition on the target speech frame to obtain text chat data; Performing word segmentation processing on the text chat data to obtain a word segmentation vocabulary; Extracting words associated with emotions from the word segmentation vocabulary as keywords; Performing emotion recognition on the keywords to obtain text emotions; Extracting features from the target speech frame to obtain target speech features; Performing emotion recognition on the voice chat data according to the target voice features to obtain voice emotions; Determining a comprehensive emotion according to the voice emotion and the text emotion; Calculating the attitude evaluation score of the candidate real estate agent based on the comprehensive emotion; Determine a viewing evaluation score based on multi-dimensional evaluation indicators, the viewing evaluation score is used to evaluate the viewing success rate of the real estate agent, and the multi-dimensional evaluation indicators at least include historical transaction rate, historical transaction price and historical unpopular property transaction rate; A target real estate agent is determined from among the candidate real estate agents according to the attitude evaluation score and the viewing evaluation score.

2. The method according to claim 1, characterized in that The step of determining a target real estate agent from among the candidate real estate agents according to the attitude evaluation score and the viewing evaluation score comprises: If the number of candidate real estate agents corresponding to the maximum value of the attitude evaluation score is 1, the candidate real estate agent corresponding to the maximum value of the attitude evaluation score is used as the target real estate agent; If the number of candidate real estate agents corresponding to the maximum value of the attitude evaluation score is greater than 1, the candidate real estate agent corresponding to the maximum value of the attitude evaluation score is used as a reference real estate agent; The reference real estate agent corresponding to the maximum viewing evaluation score is used as the target real estate agent.

3. The method according to claim 1, characterized in that The method of determining the viewing evaluation score based on the multi-dimensional evaluation indicators includes: Obtain the historical viewing quantity and historical transaction quantity of the candidate real estate agent; Obtaining a historical transaction rate of the candidate real estate agent, where the historical transaction rate is calculated based on the historical number of viewings and the historical number of transactions; The viewing evaluation score is calculated based on the historical transaction rate, the historical transaction price and the historical unpopular property transaction rate.

4. The method according to any one of claims 1 to 3, characterized in that: The matching of the viewing request with the viewing information of the real estate agent to obtain a candidate real estate agent includes: Searching for viewing information of all real estate agents from a preset real estate agent database, wherein the viewing information includes an estimated viewing time and an estimated viewing area; The real estate agent whose estimated showing time includes the desired showing time and whose estimated showing area includes the location where the desired showing house is located is determined as a candidate real estate agent.

5. A device recommended by a real estate agent, characterized in that include: A communication unit, configured to receive a viewing request sent by a target user, wherein the viewing request includes an expected viewing time, an expected viewing house, and expected preference information for a real estate agent; A processing unit, configured to match the viewing request with the viewing information of the real estate agent to obtain a candidate real estate agent; If it is determined that there is no user evaluation information, obtaining voice chat data of a user associated with the candidate real estate agent, wherein the user evaluation information includes a user rating level and a user rating content; Performing frame processing on the voice chat data to obtain a first voice frame; performing preprocessing on the first voice frame to obtain a target voice frame; performing voice recognition on the target voice frame to obtain text chat data; performing word segmentation processing on the text chat data to obtain a word segmentation vocabulary; Extracting words associated with emotions from the word segmentation vocabulary as keywords; performing emotion recognition on the keywords to obtain text emotions; performing feature extraction on the target speech frame to obtain target speech features; Performing emotion recognition on the voice chat data according to the target voice features to obtain voice emotions; determining a comprehensive emotion according to the voice emotions and the text emotions; and calculating an attitude evaluation score of the candidate real estate agent according to the comprehensive emotion; A viewing evaluation score is determined based on multi-dimensional evaluation indicators, wherein the viewing evaluation score is used to evaluate the viewing success rate of the real estate agent, and the multi-dimensional evaluation indicators at least include historical transaction rate, historical transaction price and historical unpopular property transaction rate; a target real estate agent is determined from the candidate real estate agents based on the attitude evaluation score and the viewing evaluation score.

6. A computer device, characterized in that: The method comprises a processor, a memory and a communication interface, wherein the memory stores a computer program, the computer program is configured to be executed by the processor, and the computer program includes instructions for executing the steps in the method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which enables a computer to execute to implement the method according to any one of claims 1 to 4.

Images