Public rental housing contract management method and device, computer device and storage device

Abstract

The invention belongs to the technical field of electronic information, and particularly relates to a public rental house contract management method and device, a computer device and a storage device. The method comprises the following steps: receiving user information input by a user; extracting a first user feature according to the user information; the first user feature is input into a pre-trained first machine learning model to obtain a lease judgment result, and the lease judgment result is used for representing whether the user is a tenant allowing to lease the public rental housing; and if the lease judgment result is that the user is the tenant allowed to lease the public rental housing, generating an unconfirmed lease contract according to the user information, and generating a confirmed lease contract after receiving confirmation information input by the user. According to the method, the qualification of the tenant can be audited intelligently, and the lease contract is prepared after the audit is passed without manual participation, so that the overall efficiency of contract signing is improved, and the situation of manual operation errors is avoided.

Description

Technical Field

[0001] The present invention belongs to the field of electronic information technology, and in particular relates to a public rental housing contract management method, device, computer device and storage device. Background Art

[0002] Public rental housing is a transitional solution for people with housing difficulties. Its purpose is to provide housing assistance to people who are not low-income people but have housing difficulties. Public rental housing is not owned by individuals, but by the government or public institutions. It is rented out to new employees, including fresh graduates and people who migrate to cities for work, at a price lower than the market price or at a price that the tenants can afford.

[0003] In the existing public housing rental market, the management of rental contracts is relatively complicated. It involves many operations that need to be completed manually, such as reviewing the qualifications of tenants one by one, which takes a lot of manpower and time, and carefully preparing contract documents after the review. These operations often require relevant personnel to invest a lot of energy and time. The overall process is relatively cumbersome and lengthy, and there is a risk of manual error. Summary of the invention

[0004] Based on this, the purpose of the present invention is to provide a public rental housing contract management method, device, computer device and storage device, so as to fundamentally solve the problem that the existing public rental housing lease contract requires manual review of qualifications and preparation of documents, resulting in the overall process being cumbersome and lengthy, and there is a risk of manual operation errors.

[0005] The embodiment of the present invention is implemented by providing a public rental housing contract management method, comprising the following steps:

[0006] Receive user information input by the user;

[0007] extracting a first user feature according to the user information;

[0008] Inputting the first user feature into a pre-trained first machine learning model to obtain a rental judgment result, wherein the rental judgment result is used to characterize whether the user is a tenant who is allowed to rent a public housing;

[0009] If the rental judgment result is that the user is a tenant allowed to rent public housing, an unconfirmed rental contract is generated according to the user information, and after receiving the confirmation information input by the user, a confirmed rental contract is generated.

[0010] In some embodiments, after receiving the user information input by the user, the following steps are included:

[0011] generating at least one first question information according to the user information;

[0012] receiving first answer information input by the user based on the first question information;

[0013] The step of extracting the first user feature according to the user information comprises the following steps:

[0014] A first user feature is extracted according to the user information and the first answer information.

[0015] In some embodiments, the training process of the first machine learning model comprises the following steps:

[0016] Acquire a first data set, the first data set including an initial first user feature and an initial rental judgment result, the initial first user feature and the initial rental judgment result having a corresponding relationship;

[0017] Training the first machine learning model based on the first data set to obtain the pre-trained first machine learning model includes:

[0018] Dividing the first data set into a first training set and a first test set;

[0019] Initializing the first machine learning model;

[0020] The initial first user features of the first training set are input into the initialized first machine learning model to iteratively train the first machine learning model, including inputting the initial first user features of the first training set into the first machine learning model after x times of training so as to train the first machine learning model for the x+1th time, and determining whether to terminate the iterative training based on whether the initial rental judgment results of the first test set obtained by the first machine learning model after the x+1th training meet a first preset requirement, wherein the first preset requirement includes that the accuracy of the initial rental judgment results of the first test set reaches a predetermined level, and x is an integer greater than 0.

[0021] In some embodiments, generating an unconfirmed rental contract according to the user information comprises the following steps:

[0022] Extracting a second user feature according to the user information, and extracting a public rental housing feature from a public rental housing rental database;

[0023] Inputting the second user feature and the public rental housing feature into a pre-trained second machine learning model to obtain key contract information;

[0024] An unconfirmed lease contract is generated based on the key contract information and a pre-generated contract template.

[0025] In some implementations, extracting the second user feature according to the user information comprises the following steps:

[0026] generating at least one second question information according to the user information;

[0027] receiving second answer information input by the user based on the second question information;

[0028] A second user feature is extracted according to the user information and the second answer information.

[0029] In some embodiments, the training process of the second machine learning model comprises the following steps:

[0030] Acquire a second data set, where the second data set includes initial second user characteristics and initial contract key information, and the initial second user characteristics and the initial contract key information have a corresponding relationship;

[0031] Training the second machine learning model based on the second data set to obtain the pre-trained second machine learning model includes:

[0032] dividing the second data set into a second training set and a second test set;

[0033] Initializing the second machine learning model;

[0034] Inputting the initial second user features of the second training set into the initialized second machine learning model to iteratively train the second machine learning model, including inputting the initial second user features of the second training set into the second machine learning model after x times of training so as to train the second machine learning model for the x+1th time, and determining whether to terminate the iterative training based on whether the initial contract key information of the second test set obtained by the second machine learning model after the x+1th training meets the second preset requirements, wherein the second preset requirements include that the accuracy of the initial contract key information of the second test set reaches a predetermined level, and x is an integer greater than 0.

[0035] In some embodiments, the user information includes at least one of text data, picture data, and voice data;

[0036] Based on the text data, data extraction is performed by using a bag-of-words method and / or a TF-IDF method;

[0037] Based on the image data, data extraction is performed by a convolutional neural network method;

[0038] Based on the speech data, data extraction is performed using a Mel-frequency cepstral coefficient method.

[0039] Another embodiment of the present invention further provides a public rental housing contract management device, comprising:

[0040] An information receiving unit, used for receiving user information input by a user;

[0041] A feature extraction unit, configured to extract a first user feature according to the user information;

[0042] A result acquisition unit, used for inputting the first user feature into a pre-trained first machine learning model to obtain a rental judgment result, wherein the rental judgment result is used to characterize whether the user is a tenant who is allowed to rent a public rental housing;

[0043] The contract generating unit is used to generate an unconfirmed lease contract according to the user information if the lease judgment result is that the user is a tenant allowed to lease public housing, and to generate a confirmed lease contract after receiving confirmation information input by the user.

[0044] Another embodiment of the present invention further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the public rental housing contract management method as described in any one of the embodiments of the present invention are implemented.

[0045] Another embodiment of the present invention further provides a storage device, wherein the storage device stores a computer program, and the computer program can be executed to implement the steps of the public rental housing contract management method as described in any one of the embodiments of the present invention.

[0046] The public rental housing contract management method provided by the embodiment of the present invention extracts a first user feature based on the user information after receiving the user information input by the user, and then inputs the first user feature into a pre-trained first machine learning model to obtain a rental judgment result. If the rental judgment result is that the user is a tenant who is allowed to rent a public rental housing, an unconfirmed rental contract is generated based on the user information, and a confirmed rental contract is generated after receiving the confirmation information input by the user. The public rental housing contract management method intelligently reviews the tenant's qualifications and prepares the rental contract after the review is passed, without manual participation, thereby improving the overall efficiency of contract signing and avoiding manual operation errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a flowchart of a public rental housing contract management method provided by an embodiment of the present invention;

[0048] Figure 2 It is a structural schematic diagram of a public rental housing contract management device provided by an embodiment of the present invention;

[0049] The following specific implementations will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0050] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0051] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0052] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0053] Embodiment 1

[0054] See also Figure 1 , which shows a public rental housing contract management method in a first embodiment of the present invention. For the sake of convenience, only the part related to the embodiment of the present invention is shown. The public rental housing contract management method provided by the embodiment of the present invention includes the following steps:

[0055] S10, receiving user information input by the user;

[0056] S10, extracting a first user feature according to the user information;

[0057] S30, inputting the first user feature into a pre-trained first machine learning model to obtain a rental judgment result, where the rental judgment result is used to characterize whether the user is a tenant who is allowed to rent a public rental housing;

[0058] S40: If the rental judgment result is that the user is a tenant who is allowed to rent public housing, an unconfirmed rental contract is generated according to the user information, and after receiving the confirmation information input by the user, a confirmed rental contract is generated.

[0059] The user enters user information in the client (the client can be a mobile terminal or a web terminal), the client transmits the user information to the server, and the server receives the user information entered by the user. The user information includes multiple items, for example, the user information includes the user's name, user's age, user's ID number, user's ethnicity, user's native place, user's facial scan photos, family members' names, family members' ages, family members' ID numbers, family members' ethnicity, family members' native places, family members' facial scan photos, etc.

[0060] After receiving the user information, the first user feature is extracted based on the user information. There are multiple first user features extracted, for example, the first user feature includes user name, user age, user ethnicity, user native place, user face scan photos, family members' names, family members' ages, family members' ethnicity, family members' native places, family members' face scan photos, etc.

[0061] It is worth mentioning that the above user information and the first user feature are only examples and do not limit these contents. The user information may be other, and the first user feature extracted based on the user information may also be other.

[0062] After extracting the first user feature, the first user feature is input into a pre-trained first machine learning model to obtain a rental judgment result. Through the rental judgment result, it can be determined whether the user is a tenant allowed to rent public housing, that is, whether the user is eligible to rent public housing.

[0063] If the rental judgment result is that the user is a tenant allowed to rent public housing, an unconfirmed rental contract is automatically generated based on the user information, and the rental contract is fed back to the client. The user enters confirmation information in the client, and the client transmits the confirmation information to the server. The server receives the confirmation information entered by the user and automatically generates a confirmed rental contract.

[0064] Among them, the confirmation information entered by the user in the client can be a signature, or a signature and the first installment of rent. The client transmits the confirmation information to the server. The server receives the signature, or the signature and the first installment of rent entered by the user, determines that the user has confirmed, and automatically generates a confirmed lease contract. If the confirmation information involves the first installment of rent, this public rental housing contract management method is closely combined with digital financial technology to create an extremely convenient online payment method for tenants. Tenants can easily pay rent online through a highly integrated payment gateway. The entire payment process is greatly simplified and can be completed quickly with just a few simple clicks. In addition, in terms of fund security, this public rental housing contract management method adopts multiple security protection technologies. The payment information is encrypted and transmitted through advanced encryption algorithms to prevent the payment data from being stolen or tampered with. At the same time, it cooperates with professional payment security institutions to monitor payment risks in real time, issue warnings in time and take countermeasures to ensure the safety and stability of tenant payment funds.

[0065] Among them, when automatically generating a confirmed lease contract, it can be updated based on the unconfirmed lease contract, and the updated one is the confirmed lease contract. Alternatively, a new confirmed lease contract is generated, and the new confirmed lease contract covers the unconfirmed lease contract. Compared with the two, it is simpler to automatically generate a confirmed lease contract by updating, and it can avoid processing redundant data.

[0066] If the rental judgment result is that the user is not allowed to rent public housing, the server will feedback the review rejection information to the client, and the client will display the review rejection information to the user, for example, the review rejection information is the words "not eligible".

[0067] The public rental housing contract management method provided by the embodiment of the present invention extracts a first user feature based on the user information after receiving the user information input by the user, and then inputs the first user feature into a pre-trained first machine learning model to obtain a rental judgment result. If the rental judgment result is that the user is a tenant who is allowed to rent a public rental housing, an unconfirmed rental contract is generated based on the user information, and a confirmed rental contract is generated after receiving the confirmation information input by the user. The public rental housing contract management method intelligently reviews the tenant's qualifications and prepares the rental contract after the review is passed, without manual participation, thereby improving the overall efficiency of contract signing and avoiding manual operation errors.

[0068] In some specific embodiments of the present invention, after receiving the user information input by the user, the following steps are included:

[0069] generating at least one first question information according to the user information;

[0070] Receiving first answer information input by the user based on the first question information;

[0071] Extracting a first user feature according to the user information includes the following steps:

[0072] A first user feature is extracted according to the user information and the first answer information.

[0073] In this embodiment, after receiving the user information input by the user, the server generates at least one first question information according to the user information, and transmits the first question information to the client, the client displays the first question information to the user, the user inputs the first answer information based on the first question information, the client transmits the first answer information to the server, and the server receives the first answer information. After receiving the first answer information, the server extracts the first user feature according to the user information and the first answer information.

[0074] For example, user information includes user education. After the server receives the user education input by the user, the first question information generated according to the user education is whether there is an idea to improve education or whether the user is taking courses to improve education. The server then transmits the first question information to the client, and the user enters the first answer information in the client in combination with his or her own situation. For example, user information includes user marital status. After the server receives the user marital status input by the user, if the user is unmarried, the first question information generated by the server is whether there is an idea to get married or whether there is a boyfriend / girlfriend in a relationship. The server then transmits the first question information to the client, and the user enters the first answer information in the client in combination with his or her own situation. For example, user information includes user children. After the server receives the user children input by the user, if the user is childless, the first question information generated by the server is whether there is an idea to give birth or whether the user is pregnant. The server then transmits the first question information to the client, and the user enters the first answer information in the client in combination with his or her own situation. Of course, the above-mentioned user information and first question information are only examples, and these contents are not limited. The user information and the first question information can be other, and the first answer information can also be other.

[0075] Through the public rental housing contract management method of the present invention, since the user information is incomplete or the information is not relevant information that is conducive to feature extraction, the first extended question information is intelligently generated based on the user information input by the user, and after receiving the first answer information input by the user based on the first question information, the first user feature is extracted according to the user information and the first answer information, so that the extracted first user feature is more comprehensive and accurate, and then after the first user feature is input into the first machine learning model, the most accurate rental judgment result can be output, ensuring that the user's qualification review is the most accurate.

[0076] In some specific embodiments of the present invention, the training process of the first machine learning model includes the following steps:

[0077] Acquire a first data set, the first data set including an initial first user feature and an initial lease judgment result, and the initial first user feature and the initial lease judgment result have a corresponding relationship;

[0078] Training a first machine learning model based on the first data set to obtain a pre-trained first machine learning model includes:

[0079] Divide the first data set into a first training set and a first test set;

[0080] Initialize the first machine learning model;

[0081] The initial first user features of the first training set are input into the initialized first machine learning model to iteratively train the first machine learning model, including inputting the initial first user features of the first training set into the first machine learning model after x times of training so as to train the first machine learning model for the x+1th time, and determining whether to terminate the iterative training based on whether the initial rental judgment results of the first test set obtained by the first machine learning model after the x+1th training meet the first preset requirements, wherein the first preset requirements include that the accuracy of the initial rental judgment results of the first test set reaches a predetermined level, and x is an integer greater than 0.

[0082] Based on the actual situation of previous public rental housing contract management, a first data set is generated, the first data set includes an initial first user feature and an initial rental judgment result, and the initial first user feature and the initial rental judgment result have a corresponding relationship. After obtaining the first data set, the server trains a first machine learning model based on the first data set to obtain a pre-trained first machine learning model.

[0083] The specific training process of the first machine learning model is to first divide the first data set into a first training set and a first test set, and the ratio of the first training set to the first test set can be 9:1, 8:2 or 7:3, or other. Then initialize the first machine learning model to prevent the first machine learning model from preserving previous training traces. Then input the initial first user features of the first training set into the first machine learning model that has been initialized to iteratively train the first machine learning model. The iterative training process is to input the initial first user features of the first training set into the first machine learning model after x trainings, so as to train the first machine learning model for the x+1th time, and after the x+1th training, verify the first machine learning model after the x+1th training through the first test set to determine whether the initial rental judgment result of the first test set meets the first preset requirement, and then determine whether to terminate the iterative training. If it is determined that the first preset requirement is met, the iterative training is terminated to generate the trained first machine learning model. If it is determined that the first preset requirement is not met, iterative training is then performed. The first preset requirement includes that the accuracy of the initial rental judgment result of the first test set reaches a predetermined level, and x is an integer greater than 0.

[0084] In some specific embodiments of the present invention, generating an unconfirmed lease contract according to user information includes the following steps:

[0085] Extracting a second user feature based on the user information, and extracting a public rental housing feature from a public rental housing rental database;

[0086] Input the second user features and the public rental housing features into the pre-trained second machine learning model to obtain the contract key information;

[0087] Generate unconfirmed lease contracts based on contract key information and pre-generated contract templates.

[0088] In this embodiment, after determining that the user is a tenant who is allowed to rent public housing, the second user feature is extracted according to the user information, and the second user feature may be the same as or different from the first user feature. For example, the first user feature includes the user's name, user's age, user's ID number, user's ethnicity, user's native place, family member's name, family member's age, family member's ID number, family member's ethnicity, family member's native place, etc. Among them, the second user feature and the first user feature that are the same do not need to be extracted repeatedly. And extract public housing features from the public housing rental database, which is a database that is updated and changed in real time with the actual rental situation of public housing. The public housing features may be the total number of public housing sources, the number of remaining public housing sources, the geographical location of the remaining public housing sources, and the surrounding environment information of the remaining public housing sources.

[0089] It is worth mentioning that the above-mentioned second user characteristics and public rental housing characteristics are only examples and do not limit these contents. The second user characteristics and public rental housing characteristics may be other.

[0090] After extracting the second user features and public rental housing features, the second user features and public rental housing features are input into the pre-trained second machine learning model to obtain the key contract information. For example, the key contract information includes relevant information of the selected public rental housing (relevant buildings, relevant buildings, relevant households), rent for each period, remarks, etc.

[0091] After obtaining the key contract information, an unconfirmed lease contract is generated based on the key contract information and the pre-generated contract template. There are multiple contract templates pre-generated for different situations, and the corresponding contract template is selected based on the actual situation.

[0092] The public rental housing contract management method provided by the embodiment of the present invention, after determining that the user is a tenant who is allowed to rent public rental housing, extracts the second user feature according to the user information, and extracts the public rental housing feature from the public rental housing rental database, and inputs the second user feature and the public rental housing feature into the pre-trained second machine learning model to obtain the key contract information, and finally generates an unconfirmed lease contract according to the key contract information and the pre-generated contract template. In the process of preparing the lease contract, the machine learning algorithm is involved, and it does not rely on simple rules or users to select housing sources independently. It can make more accurate decisions, such as allocating housing sources, adjusting rents, etc., and can intelligently adjust the allocation of public housing resources according to real-time data and analysis results to ensure maximum resource utilization, while effectively reducing idle resources and waste.

[0093] In some specific embodiments of the present invention, extracting the second user feature according to the user information includes the following steps:

[0094] generating at least one second question information according to the user information;

[0095] receiving second answer information input by the user based on the second question information;

[0096] The second user feature is extracted according to the user information and the second answer information.

[0097] In this embodiment, the server generates at least one second question information according to the user information, and transmits the second question information to the client, the client displays the second question information to the user, the user inputs the second answer information based on the second question information, the client transmits the second answer information to the server, and the server receives the second answer information. After receiving the second answer information, the server extracts the second user feature according to the user information and the second answer information.

[0098] The generated second question information may be the same as or different from the second question information. For example, the user information includes the user's education level. After the server receives the user's education level input by the user, the second question information generated according to the user's education level is whether there is an idea of ​​improving education level or whether the user is taking courses to improve education level. The server then transmits the second question information to the client, and the user enters the second answer information in the client in combination with his or her own situation. For example, the user information includes the user's marital status. After the server receives the user's marital status input by the user, if the user is unmarried, the second question information generated by the server is whether there is an idea of ​​getting married or whether there is a boyfriend / girlfriend in a relationship. The server then transmits the second question information to the client, and the user enters the second answer information in the client in combination with his or her own situation. For example, the user information includes the user's children. After the server receives the user's children input by the user, if the user is childless, the second question information generated by the server is whether there is an idea of ​​giving birth or whether the user is pregnant. The server then transmits the second question information to the client, and the user enters the second answer information in the client in combination with his or her own situation. For example, the user information includes the situation of co-living parents. After the server receives the situation of co-living parents input by the user, if there are no parents living together for a long time, the second question information generated by the server is whether there are any ideas of parents living together, etc. The server then transmits the second question information to the client, and the user enters the second answer information in the client based on his own situation. For example, the server determines that the user information input by the user does not include the user's expectations, and the second question information generated by the server is based on whether there are relevant expectations / ideas for house selection. The server then transmits the second question information to the client, and the user enters the second answer information in the client based on his own situation. In the case where the second question information is the same as the first question information, the second question information does not need to be generated repeatedly. Of course, the above-mentioned user information and second question information are only examples, and there is no limitation on these contents. The user information and the second question information can be other, and the second answer information can also be other.

[0099] Through the public rental housing contract management method of the present invention, since the user information is incomplete or the information is not relevant information that is conducive to feature extraction, the second extended question information is intelligently generated based on the user information input by the user, and after receiving the second answer information input by the user based on the second question information, the second user features are extracted according to the user information and the second answer information, so that the extracted second user features are more comprehensive and accurate, and then after the second user features are input into the second machine learning model, the most accurate contract key information can be output, ensuring that the intelligently generated lease contract is the most accurate.

[0100] In some specific embodiments of the present invention, the training process of the second machine learning model includes the following steps:

[0101] Acquire a second data set, the second data set including initial second user characteristics and initial contract key information, the initial second user characteristics and the initial contract key information having a corresponding relationship;

[0102] Training a second machine learning model based on the second data set to obtain a pre-trained second machine learning model includes:

[0103] dividing the second data set into a second training set and a second test set;

[0104] Initialize a second machine learning model;

[0105] The initial second user features of the second training set are input into the initialized second machine learning model to iteratively train the second machine learning model, including inputting the initial second user features of the second training set into the second machine learning model after x times of training so as to train the second machine learning model for the x+1th time, and determining whether to terminate the iterative training based on whether the initial contract key information of the second test set obtained by the second machine learning model after the x+1th training meets the second preset requirements, wherein the second preset requirements include that the accuracy rate of the initial contract key information of the second test set reaches a predetermined level, and x is an integer greater than 0.

[0106] Based on the actual situation of public rental housing contract management in the past, a second data set is generated, and the second data set includes initial second user features and initial contract key information, and the initial second user features and initial contract key information have a corresponding relationship. After obtaining the second data set, the server trains a second machine learning model based on the second data set to obtain a pre-trained second machine learning model.

[0107] The specific training process of the second machine learning model is to first divide the second data set into a second training set and a second test set, and the ratio of the second training set to the second test set can be 9:1, 8:2 or 7:3, or other. Then initialize the second machine learning model to prevent the second machine learning model from preserving previous training traces. Then input the initial second user features of the second training set into the initialized second machine learning model to iteratively train the second machine learning model. The iterative training process is to input the initial second user features of the second training set into the second machine learning model after x trainings, so as to train the second machine learning model for the x+1th time, and after the x+1th training, verify the second machine learning model after the x+1th training through the second test set to determine whether the initial contract key information of the second test set meets the second preset requirements, and then determine whether to terminate the iterative training. If it is determined that the second preset requirements are met, the iterative training is terminated to generate the trained second machine learning model. If it is determined that the second preset requirements are not met, iterative training is then performed. Among them, the second preset requirement includes that the accuracy of the initial contract key information of the second test set reaches a predetermined level, and x is an integer greater than 0.

[0108] In some specific embodiments of the present invention, the user information includes at least one of text data, picture data and voice data;

[0109] Based on text data, data extraction is performed using the bag-of-words method and / or the TF-IDF method;

[0110] Based on the image data, data extraction is performed through the convolutional neural network method;

[0111] Based on speech data, data extraction is performed using the Mel-frequency cepstral coefficient method.

[0112] After data extraction is performed based on text data, image data or voice data, the extracted data can be used to extract the first user feature and the second user feature, or to generate the first question information and the second question information.

[0113] For text data, user features are extracted through the bag-of-words method and / or the TF-IDF method. The bag-of-words (BOW) method can ignore the grammatical and word order elements of the text and regard it as a collection of several words. The appearance of each word in the document is independent, and the frequency of the words in the text is used to describe the document, and a document is represented as a one-dimensional vector. The TF-IDF (Term frequency-inverse document frequency) method is a commonly used weighting technology for information retrieval and data mining. TF is term frequency (TermFrequency), IDF is inverse document frequency index (Inverse Document Frequency), TF-IDF is proportional to the number of times a word appears in a document, and inversely proportional to the number of times the word appears in the entire language. Then, after calculating the TF-IDF value of each word in the document, and then arranging them in descending order, take the word arranged in front as the required word.

[0114] For image data, user features are extracted through the convolutional neural network method. The convolutional neural network (CNN) method is a type of feedforward neural network (Feedforward Neural Networks) that includes convolution calculations and has a deep structure. It is one of the representative algorithms of deep learning. Convolutional neural networks have representation learning capabilities and can perform shift-invariant classification on input information according to their hierarchical structure. Therefore, they are also called "Shift-Invariant Artificial Neural Networks (SIANN)".

[0115] For speech data, user features are extracted using the Mel-Frequency Cepstral Coefficients method. The Mel-Frequency Cepstral Coefficients (MFCCs) method refers to the coefficients that make up the Mel-Frequency Cepstral. They are derived from the cepstrum representation (a nonlinear "spectrum-of-a-spectrum") of the audio clip. The frequency band division of the Mel-Frequency Cepstral is equally spaced on the Mel scale, which better approximates the human auditory system than the linearly spaced bands used in the normal logarithmic cepstrum. This frequency bending can better represent sound, such as audio compression.

[0116] In addition to user information, the first answer information and the second answer information also include at least one of text data, image data and voice data, and corresponding data extraction can be performed based on the above-mentioned various methods.

[0117] Furthermore, the public rental housing contract management method of the present invention is adaptive and can adjust the data processing logic in real time according to the latest laws and regulations and changes in the lease agreement, ensuring that qualification review and document preparation always meet the current lease conditions and standards.

[0118] In some specific embodiments of the present invention, after the confirmed lease contract is generated, the following steps are included:

[0119] Receive change information, renewal information or termination information input by the user;

[0120] Change the lease contract according to the change information, renew the lease contract according to the renewal information, or terminate the lease contract according to the termination information.

[0121] In this embodiment, if the user needs to change, renew or terminate the lease contract, the user enters the change information, renewal information or termination information in the client, and the client transmits the change information, renewal information or termination information to the server. The server changes the lease contract according to the change information, renews the lease contract according to the renewal information, or terminates the lease contract according to the termination information. The change, renewal and termination operations of the lease contract are automatically processed without the user having to communicate with the staff on site, which greatly improves the efficiency of contract management.

[0122] In some specific embodiments of the present invention, after the confirmed lease contract is generated, the user information is associated with the door lock system in the public rental housing, and the water and electricity system in the public rental housing is enabled.

[0123] In this embodiment, after generating a confirmed lease contract, the server determines that the user is qualified to live in the public rental housing, and associates the user information with the door lock system in the public rental housing, such as extracting the user's face scan photo in the user information, or extracting the last six digits of the user's ID number in the user information as the unlocking information of the door lock system. And activate the water and electricity system in the public rental housing, and the user can activate the water and electricity system after entering the public rental housing.

[0124] For public rental housing, the door lock system is an extremely critical component. The door lock system can use advanced technology to achieve accurate identity recognition and convenient unlocking operations, which not only greatly improves the convenience of users entering and exiting, but also provides a strong guarantee for the safety of the house. At the same time, the water and electricity system is also one of the indispensable hardware. The water and electricity system usually includes a water and electricity controller and a water and electricity sensor, which can control and monitor the use of water and electricity in the house in real time and accurately, and promptly feedback the relevant data to the server, so that the server can effectively manage and monitor water and electricity resources, and ensure the rationality and efficiency of water and electricity use. These hardware devices complement each other and together provide a solid foundation and necessary support for the smooth operation of the entire public rental housing system, effectively ensuring that the public rental housing system can continue to operate stably and efficiently for a long time, thereby bringing great convenience and optimization to the management and service of the public rental housing system.

[0125] Embodiment 2

[0126] See also Figure 2 , is a schematic diagram of the structure of a public rental housing contract management device provided by the second embodiment of the present invention. For the convenience of description, only the part related to the embodiment of the present invention is shown. The public rental housing contract management device includes:

[0127] An information receiving unit 11, used to receive user information input by a user;

[0128] A feature extraction unit 12, configured to extract a first user feature according to the user information;

[0129] A result acquisition unit 13 is used to input the first user feature into a pre-trained first machine learning model to obtain a rental judgment result, where the rental judgment result is used to characterize whether the user is a tenant who is allowed to rent a public rental housing;

[0130] The contract generating unit 14 is used to generate an unconfirmed lease contract according to the user information if the lease judgment result is that the user is a tenant allowed to lease public housing, and to generate a confirmed lease contract after receiving confirmation information input by the user.

[0131] The user enters user information in the client (the client can be a mobile terminal or a web terminal), the client transmits the user information to the server, and the server receives the user information entered by the user. The user information includes multiple items, for example, the user information includes the user's name, user's age, user's ID number, user's ethnicity, user's native place, user's facial scan photos, family members' names, family members' ages, family members' ID numbers, family members' ethnicity, family members' native places, family members' facial scan photos, etc.

[0132] After receiving the user information, the first user feature is extracted based on the user information. There are multiple first user features extracted, for example, the first user feature includes user name, user age, user ethnicity, user native place, user face scan photos, family members' names, family members' ages, family members' ethnicity, family members' native places, family members' face scan photos, etc.

[0133] It is worth mentioning that the above user information and the first user feature are only examples and do not limit these contents. The user information may be other, and the first user feature extracted based on the user information may also be other.

[0134] After extracting the first user feature, the first user feature is input into a pre-trained first machine learning model to obtain a rental judgment result. Through the rental judgment result, it can be determined whether the user is a tenant allowed to rent public housing, that is, whether the user is eligible to rent public housing.

[0135] If the rental judgment result is that the user is a tenant allowed to rent public housing, an unconfirmed rental contract is automatically generated based on the user information, and the rental contract is fed back to the client. The user enters confirmation information in the client, and the client transmits the confirmation information to the server. The server receives the confirmation information entered by the user and automatically generates a confirmed rental contract.

[0136] Among them, the confirmation information entered by the user in the client can be a signature, or a signature and the first installment of rent. The client transmits the confirmation information to the server. The server receives the signature, or the signature and the first installment of rent entered by the user, determines that the user has made a confirmation, and automatically generates a confirmed lease contract. If the confirmation information involves the first installment of rent, this public rental housing contract management device is closely combined with digital financial technology to create an extremely convenient online payment method for tenants. Tenants can easily pay rent online through a highly integrated payment gateway. The entire payment process is greatly simplified and can be completed quickly with just a few simple clicks. In addition, in terms of fund security, this public rental housing contract management device adopts multiple security protection technologies. The payment information is encrypted and transmitted through advanced encryption algorithms to prevent the payment data from being stolen or tampered with. At the same time, it cooperates with professional payment security institutions to monitor payment risks in real time, issue warnings in time and take countermeasures to ensure the safety and stability of tenant payment funds.

[0137] Among them, when automatically generating a confirmed lease contract, it can be updated based on the unconfirmed lease contract, and the updated one is the confirmed lease contract. Alternatively, a new confirmed lease contract is generated, and the new confirmed lease contract covers the unconfirmed lease contract. Compared with the two, it is simpler to automatically generate a confirmed lease contract by updating, and it can avoid processing redundant data.

[0138] If the rental judgment result is that the user is not allowed to rent public housing, the server will feedback the review rejection information to the client, and the client will display the review rejection information to the user, for example, the review rejection information is the words "not eligible".

[0139] The public rental housing contract management device provided by the embodiment of the present invention extracts a first user feature based on the user information after receiving the user information input by the user, and then inputs the first user feature into a pre-trained first machine learning model to obtain a rental judgment result. If the rental judgment result is that the user is a tenant who is allowed to rent a public rental housing, an unconfirmed rental contract is generated based on the user information, and a confirmed rental contract is generated after receiving the confirmation information input by the user. The public rental housing contract management device intelligently reviews the qualifications of tenants and prepares a rental contract after the review is passed, without the need for manual participation, thereby improving the overall efficiency of contract signing and avoiding manual operation errors.

[0140] In some specific embodiments of the present invention, the public rental housing contract management device includes:

[0141] A first question generating unit, configured to generate at least one first question information according to user information;

[0142] A first answer receiving unit, configured to receive first answer information input by a user based on the first question information;

[0143] The feature extraction unit 12 comprises:

[0144] The first feature extraction module is used to extract the first user feature according to the user information and the first answer information.

[0145] In this embodiment, after receiving the user information input by the user, the server generates at least one first question information according to the user information, and transmits the first question information to the client, the client displays the first question information to the user, the user inputs the first answer information based on the first question information, the client transmits the first answer information to the server, and the server receives the first answer information. After receiving the first answer information, the server extracts the first user feature according to the user information and the first answer information.

[0146] For example, user information includes user education. After the server receives the user education input by the user, the first question information generated according to the user education is whether there is an idea to improve education or whether the user is taking courses to improve education. The server then transmits the first question information to the client, and the user enters the first answer information in the client in combination with his or her own situation. For example, user information includes user marital status. After the server receives the user marital status input by the user, if the user is unmarried, the first question information generated by the server is whether there is an idea to get married or whether there is a boyfriend / girlfriend in a relationship. The server then transmits the first question information to the client, and the user enters the first answer information in the client in combination with his or her own situation. For example, user information includes user children. After the server receives the user children input by the user, if the user is childless, the first question information generated by the server is whether there is an idea to give birth or whether the user is pregnant. The server then transmits the first question information to the client, and the user enters the first answer information in the client in combination with his or her own situation. Of course, the above-mentioned user information and first question information are only examples, and these contents are not limited. The user information and the first question information can be other, and the first answer information can also be other.

[0147] Through the public rental housing contract management device of the present invention, since the user information is incomplete or the information is not relevant information that is conducive to feature extraction, the first extended question information is intelligently generated based on the user information input by the user, and after receiving the first answer information input by the user based on the first question information, the first user feature is extracted according to the user information and the first answer information, so that the extracted first user feature is more comprehensive and accurate, and then after the first user feature is input into the first machine learning model, the most accurate rental judgment result can be output to ensure the most accurate user qualification review.

[0148] In some specific embodiments of the present invention, the public rental housing contract management device includes:

[0149] A first data set acquisition unit, configured to acquire a first data set, wherein the first data set includes an initial first user feature and an initial rental judgment result, and the initial first user feature and the initial rental judgment result have a corresponding relationship;

[0150] The first training unit is used to train a first machine learning model based on a first data set to obtain a pre-trained first machine learning model, including:

[0151] Divide the first data set into a first training set and a first test set;

[0152] Initialize the first machine learning model;

[0153] The initial first user features of the first training set are input into the initialized first machine learning model to iteratively train the first machine learning model, including inputting the initial first user features of the first training set into the first machine learning model after x times of training so as to train the first machine learning model for the x+1th time, and determining whether to terminate the iterative training based on whether the initial rental judgment results of the first test set obtained by the first machine learning model after the x+1th training meet the first preset requirements, wherein the first preset requirements include that the accuracy of the initial rental judgment results of the first test set reaches a predetermined level, and x is an integer greater than 0.

[0154] Based on the actual situation of previous public rental housing contract management, a first data set is generated, the first data set includes an initial first user feature and an initial rental judgment result, and the initial first user feature and the initial rental judgment result have a corresponding relationship. After obtaining the first data set, the server trains a first machine learning model based on the first data set to obtain a pre-trained first machine learning model.

[0155] The specific training process of the first machine learning model is to first divide the first data set into a first training set and a first test set, and the ratio of the first training set to the first test set can be 9:1, 8:2 or 7:3, or other. Then initialize the first machine learning model to prevent the first machine learning model from preserving previous training traces. Then input the initial first user features of the first training set into the first machine learning model that has been initialized to iteratively train the first machine learning model. The iterative training process is to input the initial first user features of the first training set into the first machine learning model after x trainings, so as to train the first machine learning model for the x+1th time, and after the x+1th training, verify the first machine learning model after the x+1th training through the first test set to determine whether the initial rental judgment result of the first test set meets the first preset requirement, and then determine whether to terminate the iterative training. If it is determined that the first preset requirement is met, the iterative training is terminated to generate the trained first machine learning model. If it is determined that the first preset requirement is not met, iterative training is then performed. The first preset requirement includes that the accuracy of the initial rental judgment result of the first test set reaches a predetermined level, and x is an integer greater than 0.

[0156] In some specific embodiments of the present invention, the contract generation unit 14 includes:

[0157] A second feature extraction module is used to extract a second user feature according to the user information, and to extract a public rental housing feature from a public rental housing rental database;

[0158] An information acquisition module, used for inputting the second user characteristics and the public rental housing characteristics into a pre-trained second machine learning model to obtain key contract information;

[0159] The contract generation module is used to generate an unconfirmed lease contract based on key contract information and a pre-generated contract template.

[0160] In this embodiment, after determining that the user is a tenant who is allowed to rent public housing, the second user feature is extracted according to the user information, and the second user feature may be the same as or different from the first user feature. For example, the first user feature includes the user's name, user's age, user's ID number, user's ethnicity, user's native place, family member's name, family member's age, family member's ID number, family member's ethnicity, family member's native place, etc. Among them, the second user feature and the first user feature that are the same do not need to be extracted repeatedly. And extract public housing features from the public housing rental database, which is a database that is updated and changed in real time with the actual rental situation of public housing. The public housing features may be the total number of public housing sources, the number of remaining public housing sources, the geographical location of the remaining public housing sources, and the surrounding environment information of the remaining public housing sources.

[0161] It is worth mentioning that the above-mentioned second user characteristics and public rental housing characteristics are only examples and do not limit these contents. The second user characteristics and public rental housing characteristics may be other.

[0162] After extracting the second user features and public rental housing features, the second user features and public rental housing features are input into the pre-trained second machine learning model to obtain the key contract information. For example, the key contract information includes relevant information of the selected public rental housing (relevant buildings, relevant buildings, relevant households), rent for each period, remarks, etc.

[0163] After obtaining the key contract information, an unconfirmed lease contract is generated based on the key contract information and the pre-generated contract template. There are multiple contract templates pre-generated for different situations, and the corresponding contract template is selected based on the actual situation.

[0164] The public rental housing contract management device provided by the embodiment of the present invention, after determining that the user is a tenant who is allowed to rent public rental housing, extracts the second user feature according to the user information, and extracts the public rental housing feature from the public rental housing rental database, and inputs the second user feature and the public rental housing feature into the pre-trained second machine learning model to obtain the key contract information, and finally generates an unconfirmed lease contract according to the key contract information and the pre-generated contract template. In the process of preparing the lease contract, the machine learning algorithm is involved, and it does not rely on simple rules or users to select housing sources independently. It can make more accurate decisions, such as allocating housing sources, adjusting rents, etc., and can intelligently adjust the allocation of public housing resources according to real-time data and analysis results to ensure maximum resource utilization, while effectively reducing idle resources and waste.

[0165] In some specific embodiments of the present invention, the second feature extraction module includes:

[0166] A second question generating submodule, used to generate at least one second question information according to the user information;

[0167] A second answer receiving submodule, used for receiving second answer information input by the user based on the second question information;

[0168] The second feature extraction submodule is used to extract the second user feature according to the user information and the second answer information.

[0169] In this embodiment, the server generates at least one second question information according to the user information, and transmits the second question information to the client, the client displays the second question information to the user, the user inputs the second answer information based on the second question information, the client transmits the second answer information to the server, and the server receives the second answer information. After receiving the second answer information, the server extracts the second user feature according to the user information and the second answer information.

[0170] The generated second question information may be the same as or different from the second question information. For example, the user information includes the user's education level. After the server receives the user's education level input by the user, the second question information generated according to the user's education level is whether there is an idea of ​​improving education level or whether the user is taking courses to improve education level. The server then transmits the second question information to the client, and the user enters the second answer information in the client in combination with his or her own situation. For example, the user information includes the user's marital status. After the server receives the user's marital status input by the user, if the user is unmarried, the second question information generated by the server is whether there is an idea of ​​getting married or whether there is a boyfriend / girlfriend in a relationship. The server then transmits the second question information to the client, and the user enters the second answer information in the client in combination with his or her own situation. For example, the user information includes the user's children. After the server receives the user's children input by the user, if the user is childless, the second question information generated by the server is whether there is an idea of ​​giving birth or whether the user is pregnant. The server then transmits the second question information to the client, and the user enters the second answer information in the client in combination with his or her own situation. For example, the user information includes the situation of co-living parents. After the server receives the situation of co-living parents input by the user, if there are no parents living together for a long time, the second question information generated by the server is whether there are any ideas of parents living together, etc. The server then transmits the second question information to the client, and the user enters the second answer information in the client based on his own situation. For example, the server determines that the user information input by the user does not include the user's expectations, and the second question information generated by the server is based on whether there are relevant expectations / ideas for house selection. The server then transmits the second question information to the client, and the user enters the second answer information in the client based on his own situation. In the case where the second question information is the same as the first question information, the second question information does not need to be generated repeatedly. Of course, the above-mentioned user information and second question information are only examples, and there is no limitation on these contents. The user information and the second question information can be other, and the second answer information can also be other.

[0171] Through the public rental housing contract management device of the present invention, since the user information is incomplete or the information is not relevant information that is conducive to feature extraction, the relevant extended second question information is intelligently generated based on the user information input by the user, and after receiving the second answer information input by the user based on the second question information, the second user features are extracted according to the user information and the second answer information, so that the extracted second user features are more comprehensive and accurate. After the second user features are input into the second machine learning model, the most accurate contract key information can be output, ensuring that the intelligently generated lease contract is the most accurate.

[0172] In some specific embodiments of the present invention, the public rental housing contract management device includes:

[0173] A second data set acquisition unit, used to acquire a second data set, the second data set including initial second user characteristics and initial contract key information, the initial second user characteristics and the initial contract key information having a corresponding relationship;

[0174] The second training unit is used to train a second machine learning model based on the second data set to obtain a pre-trained second machine learning model, including:

[0175] dividing the second data set into a second training set and a second test set;

[0176] Initialize the second machine learning model;

[0177] The initial second user features of the second training set are input into the initialized second machine learning model to iteratively train the second machine learning model, including inputting the initial second user features of the second training set into the second machine learning model after x times of training so as to train the second machine learning model for the x+1th time, and determining whether to terminate the iterative training based on whether the initial contract key information of the second test set obtained by the second machine learning model after the x+1th training meets the second preset requirements, wherein the second preset requirements include that the accuracy rate of the initial contract key information of the second test set reaches a predetermined level, and x is an integer greater than 0.

[0178] Based on the actual situation of public rental housing contract management in the past, a second data set is generated, and the second data set includes initial second user features and initial contract key information, and the initial second user features and initial contract key information have a corresponding relationship. After obtaining the second data set, the server trains a second machine learning model based on the second data set to obtain a pre-trained second machine learning model.

[0179] The specific training process of the second machine learning model is to first divide the second data set into a second training set and a second test set, and the ratio of the second training set to the second test set can be 9:1, 8:2 or 7:3, or other. Then initialize the second machine learning model to prevent the second machine learning model from preserving previous training traces. Then input the initial second user features of the second training set into the initialized second machine learning model to iteratively train the second machine learning model. The iterative training process is to input the initial second user features of the second training set into the second machine learning model after x trainings, so as to train the second machine learning model for the x+1th time, and after the x+1th training, verify the second machine learning model after the x+1th training through the second test set to determine whether the initial contract key information of the second test set meets the second preset requirements, and then determine whether to terminate the iterative training. If it is determined that the second preset requirements are met, the iterative training is terminated to generate the trained second machine learning model. If it is determined that the second preset requirements are not met, iterative training is then performed. Among them, the second preset requirement includes that the accuracy of the initial contract key information of the second test set reaches a predetermined level, and x is an integer greater than 0.

[0180] In some specific embodiments of the present invention, the user information includes at least one of text data, picture data and voice data;

[0181] The public rental housing contract management device includes:

[0182] A first data extraction unit, configured to extract data based on text data by using a bag-of-words method and / or a TF-IDF method;

[0183] A second data extraction unit, used for extracting data based on the image data by a convolutional neural network method;

[0184] The third data extraction unit is used to extract data based on the speech data by using the Mel-frequency cepstral coefficient method.

[0185] After data extraction is performed based on text data, image data or voice data, the extracted data can be used to extract the first user feature and the second user feature, or to generate the first question information and the second question information.

[0186] For text data, user features are extracted through the bag-of-words method and / or the TF-IDF method. The bag-of-words (BOW) method can ignore the grammatical and word order elements of the text and regard it as a collection of several words. The appearance of each word in the document is independent, and the frequency of the words in the text is used to describe the document, and a document is represented as a one-dimensional vector. The TF-IDF (Term frequency-inverse document frequency) method is a commonly used weighting technology for information retrieval and data mining. TF is term frequency (TermFrequency), IDF is inverse document frequency index (Inverse Document Frequency), TF-IDF is proportional to the number of times a word appears in a document, and inversely proportional to the number of times the word appears in the entire language. Then, after calculating the TF-IDF value of each word in the document, and then arranging them in descending order, take the word arranged in front as the required word.

[0187] For image data, user features are extracted through the convolutional neural network method. The convolutional neural network (CNN) method is a type of feedforward neural network (Feedforward Neural Networks) that includes convolution calculations and has a deep structure. It is one of the representative algorithms of deep learning. Convolutional neural networks have representation learning capabilities and can perform shift-invariant classification on input information according to their hierarchical structure. Therefore, they are also called "Shift-Invariant Artificial Neural Networks (SIANN)".

[0188] For speech data, user features are extracted using the Mel-Frequency Cepstral Coefficients method. The Mel-Frequency Cepstral Coefficients (MFCCs) method refers to the coefficients that make up the Mel-Frequency Cepstral. They are derived from the cepstrum representation (a nonlinear "spectrum-of-a-spectrum") of the audio clip. The frequency band division of the Mel-Frequency Cepstral is equally spaced on the Mel scale, which better approximates the human auditory system than the linearly spaced bands used in the normal logarithmic cepstrum. This frequency bending can better represent sound, such as audio compression.

[0189] In addition to user information, the first answer information and the second answer information also include at least one of text data, image data and voice data, and corresponding data extraction can be performed based on the above-mentioned various methods.

[0190] Furthermore, the public rental housing contract management device of the present invention is adaptive and can adjust the data processing logic in real time according to the latest laws and regulations and changes in the lease agreement, ensuring that qualification review and document preparation always meet the current lease conditions and standards.

[0191] In some specific embodiments of the present invention, the public rental housing contract management device includes:

[0192] A second information receiving unit, used to receive change information, renewal information or termination information input by a user;

[0193] The contract processing unit is used to change the lease contract according to the change information, renew the lease contract according to the renewal information, or terminate the lease contract according to the termination information.

[0194] In this embodiment, if the user needs to change, renew or terminate the lease contract, the user enters the change information, renewal information or termination information in the client, and the client transmits the change information, renewal information or termination information to the server. The server changes the lease contract according to the change information, renews the lease contract according to the renewal information, or terminates the lease contract according to the termination information. The change, renewal and termination operations of the lease contract are automatically processed without the user having to communicate with the staff on site, which greatly improves the efficiency of contract management.

[0195] In some specific embodiments of the present invention, in the contract generation unit 14, after the confirmed lease contract is generated, the user information and the door lock system in the public rental housing are associated, and the water and electricity system in the public rental housing is enabled.

[0196] In this embodiment, after generating a confirmed lease contract, the server determines that the user is qualified to live in the public rental housing, and associates the user information with the door lock system in the public rental housing, such as extracting the user's face scan photo in the user information, or extracting the last six digits of the user's ID number in the user information as the unlocking information of the door lock system. And activate the water and electricity system in the public rental housing, and the user can activate the water and electricity system after entering the public rental housing.

[0197] For public rental housing, the door lock system is an extremely critical component. The door lock system can use advanced technology to achieve accurate identity recognition and convenient unlocking operations, which not only greatly improves the convenience of users entering and exiting, but also provides a strong guarantee for the safety of the house. At the same time, the water and electricity system is also one of the indispensable hardware. The water and electricity system usually includes a water and electricity controller and a water and electricity sensor, which can control and monitor the use of water and electricity in the house in real time and accurately, and promptly feedback the relevant data to the server, so that the server can effectively manage and monitor water and electricity resources, and ensure the rationality and efficiency of water and electricity use. These hardware devices complement each other and together provide a solid foundation and necessary support for the smooth operation of the entire public rental housing system, effectively ensuring that the public rental housing system can continue to operate stably and efficiently for a long time, thereby bringing great convenience and optimization to the management and service of the public rental housing system.

[0198] An embodiment of the present invention provides a computer device, which includes a processor, and the processor is used to implement the steps of the public rental housing contract management method as described above when executing a computer program stored in a memory.

[0199] The embodiment of the present invention further provides a computer-readable storage medium having a computer program (instructions) stored thereon, wherein the computer program (instructions) implements the steps of the public rental housing contract management method as described above when executed by a processor.

[0200] Exemplarily, the computer program can be divided into one or more modules, one or more modules are stored in the memory and executed by the processor to complete the present invention. One or more modules can be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the computer device. For example, the computer program can be divided into the steps of the public rental housing contract management method provided by each of the above method embodiments.

[0201] Those skilled in the art will appreciate that the above description of the computer device is merely an example and does not constitute a limitation on the computer device. The computer device may include more or fewer components than described above, or a combination of certain components, or different components, such as input and output devices, network access devices, buses, etc.

[0202] The processor may be a central processing unit (CPU), 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. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the computer device, and uses various interfaces and lines to connect various parts of the entire computer device.

[0203] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the computer device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as an interface display function, an interface interaction function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as a map interface, a selection interface, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0204] If the module / unit integrated in the computer device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, electrical signal and software distribution medium, etc.

[0205] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0206] The above-described embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the attached claims.

Claims

1. A public rental housing contract management method, characterized in that: The following steps are involved: Receive user information input by the user; extracting a first user feature according to the user information; Inputting the first user feature into a pre-trained first machine learning model to obtain a rental judgment result, wherein the rental judgment result is used to characterize whether the user is a tenant who is allowed to rent a public housing; If the rental judgment result is that the user is a tenant allowed to rent public housing, an unconfirmed rental contract is generated according to the user information, and after receiving the confirmation information input by the user, a confirmed rental contract is generated.

2. The public rental housing contract management method according to claim 1, characterized in that: After receiving the user information input by the user, the following steps are included: generating at least one first question information according to the user information; receiving first answer information input by the user based on the first question information; The step of extracting the first user feature according to the user information comprises the following steps: A first user feature is extracted according to the user information and the first answer information.

3. The public rental housing contract management method according to claim 1 or 2, characterized in that: The training process of the first machine learning model includes the following steps: Acquire a first data set, the first data set including an initial first user feature and an initial rental judgment result, the initial first user feature and the initial rental judgment result having a corresponding relationship; Training the first machine learning model based on the first data set to obtain the pre-trained first machine learning model includes: Dividing the first data set into a first training set and a first test set; Initializing the first machine learning model; The initial first user features of the first training set are input into the initialized first machine learning model to iteratively train the first machine learning model, including inputting the initial first user features of the first training set into the first machine learning model after x times of training so as to train the first machine learning model for the x+1th time, and determining whether to terminate the iterative training based on whether the initial rental judgment results of the first test set obtained by the first machine learning model after the x+1th training meet a first preset requirement, wherein the first preset requirement includes that the accuracy of the initial rental judgment results of the first test set reaches a predetermined level, and x is an integer greater than 0.

4. The public rental housing contract management method according to claim 1, characterized in that: The step of generating an unconfirmed lease contract according to the user information comprises the following steps: Extracting a second user feature according to the user information, and extracting a public rental housing feature from a public rental housing rental database; Inputting the second user feature and the public rental housing feature into a pre-trained second machine learning model to obtain key contract information; An unconfirmed lease contract is generated based on the key contract information and a pre-generated contract template.

5. The public rental housing contract management method according to claim 4 is characterized in that: The step of extracting the second user feature according to the user information comprises the following steps: generating at least one second question information according to the user information; receiving second answer information input by the user based on the second question information; A second user feature is extracted according to the user information and the second answer information.

6. The public rental housing contract management method according to claim 4 or 5, characterized in that: The training process of the second machine learning model includes the following steps: Acquire a second data set, where the second data set includes initial second user characteristics and initial contract key information, and the initial second user characteristics and the initial contract key information have a corresponding relationship; Training the second machine learning model based on the second data set to obtain the pre-trained second machine learning model includes: dividing the second data set into a second training set and a second test set; Initializing the second machine learning model; Inputting the initial second user features of the second training set into the initialized second machine learning model to iteratively train the second machine learning model, including inputting the initial second user features of the second training set into the second machine learning model after x times of training so as to train the second machine learning model for the x+1th time, and determining whether to terminate the iterative training based on whether the initial contract key information of the second test set obtained by the second machine learning model after the x+1th training meets the second preset requirements, wherein the second preset requirements include that the accuracy of the initial contract key information of the second test set reaches a predetermined level, and x is an integer greater than 0.

7. The public rental housing contract management method according to claim 1, characterized in that: The user information includes at least one of text data, picture data and voice data; Based on the text data, data extraction is performed by using a bag-of-words method and / or a TF-IDF method; Based on the image data, data extraction is performed by a convolutional neural network method; Based on the speech data, data extraction is performed using a Mel-frequency cepstral coefficient method.

8. A public rental housing contract management device, characterized in that: include: An information receiving unit, used for receiving user information input by a user; A feature extraction unit, configured to extract a first user feature according to the user information; A result acquisition unit, used for inputting the first user feature into a pre-trained first machine learning model to obtain a rental judgment result, wherein the rental judgment result is used to characterize whether the user is a tenant who is allowed to rent a public rental housing; The contract generating unit is used to generate an unconfirmed lease contract according to the user information if the lease judgment result is that the user is a tenant allowed to lease public housing, and to generate a confirmed lease contract after receiving confirmation information input by the user.

9. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the public rental housing contract management method as claimed in any one of claims 1 to 7 when executing the computer program.

10. A storage device, characterized in that: The storage device stores a computer program, which can be executed to implement the steps of the public rental housing contract management method as described in any one of claims 1 to 7.

Images