CN109360072B - Insurance product recommendation method and device, computer equipment and storage medium - Google Patents
Insurance product recommendation method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109360072B CN109360072B CN201811347747.9A CN201811347747A CN109360072B CN 109360072 B CN109360072 B CN 109360072B CN 201811347747 A CN201811347747 A CN 201811347747A CN 109360072 B CN109360072 B CN 109360072B
- Authority
- CN
- China
- Prior art keywords
- insurance product
- character
- label
- user
- music
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Finance (AREA)
- Theoretical Computer Science (AREA)
- Accounting & Taxation (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Technology Law (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The embodiment of the invention discloses an insurance product recommendation method, an insurance product recommendation device, computer equipment and a storage medium, which comprise the following steps: acquiring music activity data of a user; inputting the characteristics of the music activity data into a preset character recognition neural network model to obtain character labels of the users; and selecting a target insurance product matched with the character label from a preset insurance product database, and pushing the target insurance product to the webpage in which the user logs. Because the music data can truly reflect the character characteristics of the user, on one hand, the insurance products can be recommended in a targeted manner, and the purchase conversion rate of the insurance products is improved; on the other hand, for the user who has the requirement of purchasing insurance, the insurance product meeting the requirement of the user can be timely obtained, the process is convenient, and the energy of the user is further saved.
Description
Technical Field
The embodiment of the invention relates to the field of finance, in particular to an insurance product recommending method, an insurance product recommending device, computer equipment and a storage medium.
Background
With the development of internet technology, more and more industries adopt the form of internet, such as microblog, music, shopping, communication, financial products and the like, and great convenience is provided for life of people. Among other things, financial products include various financial instruments, such as insurance products.
Insurance is a tool for planning life finance, is a basic means of risk management under market economy conditions, and is an important support for finance and social security. With the gradual enhancement of financial consciousness of people, insurance becomes a common financial means. In the prior art, when a user purchases an insurance product, the user needs to consult an insurance agent to select the insurance product suitable for the user, and the process is complex. Meanwhile, when an insurance company recommends an insurance product to a user, the insurance company does not have an explicit target user, so that the recommended conversion rate is low.
Disclosure of Invention
The embodiment of the invention provides an insurance product recommendation method, an insurance product recommendation device, computer equipment and a storage medium.
In order to solve the technical problems, the embodiment of the invention adopts the following technical scheme: there is provided an insurance product recommendation method including the steps of:
acquiring music activity data of a user;
inputting the characteristics of the music activity data into a preset character recognition neural network model to obtain character labels of the users;
and selecting a target insurance product matched with the character label from a preset insurance product database, and pushing the target insurance product to the webpage in which the user logs.
Optionally, each insurance product in the insurance product database contains character labels of applicable groups; the selecting the target insurance product matched with the character label from a preset insurance product database comprises the following steps:
comparing the character label of the user with the label of the security product in the security product database;
and when the comparison is consistent, taking the insurance product pointed by the label which is consistent in comparison as the target insurance product.
Optionally, when the character label of the user and the label of the insurance product are multiple, the selecting the target insurance product matched with the character label from the preset insurance product database includes:
comparing the character label of the user with the label of the security product in the security product database;
when the character labels of the users and the labels of the insurance products in the insurance product database are the same, determining the number of the labels which are the same as the character labels of the users in each insurance product;
and taking the insurance product with the largest label number as the target insurance product.
Optionally, before inputting the characteristics of the music activity data into a preset character recognition neural network model to obtain the character label of the user, the method further comprises
And extracting the characteristics of the music activity data through preset characteristic extraction software.
Optionally, before inputting the music data into a preset character recognition neural network model to obtain the character label of the user, the method further includes:
acquiring music sample data marked with character labels;
training a preset convolutional neural network model according to the music sample data to obtain a character recognition neural network model for recognizing the character label.
Optionally, the training the preset convolutional neural network model according to the music sample data includes:
extracting music characteristics from the music sample data, and calculating expected values of the music characteristics in each character label;
inputting the music characteristics into a preset convolutional neural network model to obtain excitation values of each character label in the convolutional neural network model;
comparing whether the distance between the expected value and the excitation value of each character label is smaller than or equal to a preset threshold value, and updating the weight in the convolutional neural network model through a reverse algorithm repeatedly and circularly when the distance between the expected value and the excitation value is larger than the threshold value, and ending when the distance between the expected value and the excitation value is smaller than or equal to a preset first threshold value.
In order to solve the above technical problems, an embodiment of the present invention further provides an insurance product recommendation device, including:
the acquisition module is used for acquiring music activity data of the user;
the processing module is used for inputting the characteristics of the music activity data into a preset character recognition neural network model to obtain character labels of the users;
and the execution module is used for selecting a target insurance product matched with the character label from a preset insurance product database and pushing the target insurance product to the webpage in which the user logs.
Optionally, each insurance product in the insurance product database contains character labels of applicable groups; the execution module comprises:
the first processing sub-module is used for comparing the character label of the user with the label of the security product in the security product database;
and the first execution sub-module is used for taking the insurance product pointed by the label which is consistent in comparison as the target insurance product when the comparison is consistent.
Optionally, when the character tag of the user and the tag of the insurance product are multiple, the execution module includes:
the second processing sub-module is used for comparing the character label of the user with the label of the security product in the security product database;
a second execution sub-module, configured to determine, when the user's character tag and the insurance product tag in the insurance product database have the same tag, the number of tags in each insurance product that is the same as the user's character tag;
and the third execution sub-module is used for taking the insurance product with the largest label number as the target insurance product.
Optionally, the method further comprises:
and the third processing sub-module is used for extracting the characteristics of the music activity data through preset characteristic extraction software.
Optionally, the method further comprises:
the first acquisition submodule is used for acquiring music sample data marked with character labels;
and the fourth processing sub-module is used for training a preset convolutional neural network model according to the music sample data to obtain a character recognition neural network model for recognizing the character label.
Optionally, the fourth processing sub-module includes:
the second acquisition submodule is used for extracting music characteristics from the music sample data and calculating expected values of the music characteristics in each character label;
a fifth processing sub-module, configured to input the music feature into a preset convolutional neural network model, to obtain an excitation value of each character label in the convolutional neural network model;
and the fourth execution sub-module is used for comparing whether the distance between the expected value and the excitation value of each character label is smaller than or equal to a preset threshold value, and updating the weight in the convolutional neural network model through a reverse algorithm in a repeated loop iteration mode when the distance between the expected value and the excitation value is larger than the threshold value, and ending when the distance between the expected value and the excitation value is smaller than or equal to a preset first threshold value.
In order to solve the above technical problem, an embodiment of the present invention further provides a computer device, including a memory and a processor, where the memory stores computer readable instructions, and when the computer readable instructions are executed by the processor, the processor is caused to execute the steps of the insurance product recommendation method.
To solve the above technical problem, an embodiment of the present invention further provides a storage medium storing computer readable instructions, where the computer readable instructions when executed by one or more processors cause the one or more processors to execute the steps of the insurance product recommendation method.
The embodiment of the invention has the beneficial effects that: the characteristics of the music activity data of the user are input into a preset character recognition neural network model to obtain character labels of the user, and insurance products are recommended according to the character labels of the user, and the characteristics of the user can be truly reflected by the music data, so that the insurance products can be recommended in a targeted manner on one hand, and the purchase conversion rate of the insurance products is improved; on the other hand, for the user who has the requirement of purchasing insurance, the insurance product meeting the requirement of the user can be timely obtained, the process is convenient, and the energy of the user is further saved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a basic flow chart of an insurance product recommendation method according to an embodiment of the present invention;
FIG. 2 is a basic flow chart of a method for selecting a target insurance product matching with a character tag from a preset insurance product database according to an embodiment of the present invention;
FIG. 3 is a basic flow chart of a method for selecting a target insurance product matching with the character tag from a preset insurance product database according to an embodiment of the present invention;
fig. 4 is a basic flow diagram of a training method of a personality neural network model according to an embodiment of the present invention;
fig. 5 is a basic flow chart of a method for training a preset convolutional neural network model according to music sample data according to an embodiment of the present invention;
FIG. 6 is a basic block diagram of an insurance product recommendation device according to an embodiment of the present invention;
FIG. 7 is a basic block diagram of a computer device according to an embodiment of the present invention.
Detailed Description
In order to enable those skilled in the art to better understand the present invention, the following description will make clear and complete descriptions of the technical solutions according to the embodiments of the present invention with reference to the accompanying drawings.
In some of the flows described in the specification and claims of the present invention and in the foregoing figures, a plurality of operations occurring in a particular order are included, but it should be understood that the operations may be performed out of order or performed in parallel, with the order of operations such as 101, 102, etc., being merely used to distinguish between the various operations, the order of the operations themselves not representing any order of execution. In addition, the flows may include more or fewer operations, and the operations may be performed sequentially or in parallel. It should be noted that, the descriptions of "first" and "second" herein are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, and are not limited to the "first" and the "second" being different types.
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to fall within the scope of the invention.
Examples
As used herein, a "terminal" includes both a device of a wireless signal receiver having no transmitting capability and a device of receiving and transmitting hardware having receiving and transmitting hardware capable of performing bi-directional communications over a bi-directional communication link, as will be appreciated by those skilled in the art. Such a device may include: a cellular or other communication device having a single-line display or a multi-line display or a cellular or other communication device without a multi-line display; a PCS (Personal Communications Service, personal communication system) that may combine voice, data processing, facsimile and/or data communication capabilities; a PDA (Personal Digital Assistant ) that can include a radio frequency receiver, pager, internet/intranet access, web browser, notepad, calendar and/or GPS (Global Positioning System ) receiver; a conventional laptop and/or palmtop computer or other appliance that has and/or includes a radio frequency receiver. As used herein, "terminal," "terminal device" may be portable, transportable, installed in a vehicle (aeronautical, maritime, and/or land-based), or adapted and/or configured to operate locally and/or in a distributed fashion, to operate at any other location(s) on earth and/or in space. The "terminal" and "terminal device" used herein may also be a communication terminal, a network access terminal, and a music/video playing terminal, for example, may be a PDA, a MID (Mobile Internet Device ), and/or a mobile phone with a music/video playing function, and may also be a smart tv, a set top box, and other devices.
The client terminal in this embodiment is the above-described terminal.
Specifically, referring to fig. 1, fig. 1 is a basic flow chart of an insurance product recommendation method according to the present embodiment.
As shown in fig. 1, the insurance product recommendation method includes the steps of:
s1100, acquiring music activity data of a user;
music activity data is music that users often listen to or download on various platforms, such as music software, including audio files in multiple formats, e.g., MP3 s, WAVE, WMA, VQF, MIDI, AIFF, MPEG, etc.
S1200, inputting the characteristics of the music activity data into a preset character recognition neural network model to obtain character labels of users;
the music activity data is subjected to feature extraction through preset software, for example, audio in the music activity data is input into the software to obtain a spectrogram of the audio, for example, PC Sound Spectrum software, FFT spectrum analysis software, smartLive software and the like. In practical applications, in order to make frequencies in the frequency spectrum continuous and clear, pre-emphasis, windowing and fourier transform processing are usually performed on the audio to be evaluated in the process of converting the frequency spectrum.
In the embodiment of the invention, the audio file is extracted by softmax software. Wherein the extracted feature is a frequency conversion map.
The character recognition neural network model is a model trained in advance by music sample data marked with character features. Because the style of music liked by users of different character features is different, the character of the user can be analyzed by analyzing the music that the user frequently listens to or downloads, for example, the outward-character comparison user prefers to listen to cheerful and dynamic music, and the inward-character comparison user prefers to listen to music that is more lyrics and soft. And inputting the music characteristics, namely the audio frequency spectrum, into the character neural network model to obtain character characteristics represented by the music.
It should be noted that, the character features may be divided into divisions according to various existing division rules, for example, may be divided into: liveness, strength, perfection and flatness.
S1300, selecting a target insurance product matched with the character label from a preset insurance product database, and pushing the target insurance product to a webpage in which a user logs in.
The target insurance product is an insurance product recommended to the user in the embodiment of the invention, wherein the target insurance product can comprise various types of insurance, such as serious insurance, car insurance, life insurance, child insurance, travel insurance, return insurance, delay insurance and the like. Each insurance product is suitable for people with different character characteristics, for example, people with perfect characters are sensitive, easy to sacrifice, emotion is not easy to release, risks of suffering from diseases are high, and the insurance product is suitable for serious risks and can be pushed to users with perfect characters.
According to the insurance product recommendation method, the characteristics of the music activity data of the user are input into the preset character recognition neural network model to obtain the character label of the user, and the insurance product is recommended according to the character label of the user, and the characteristics of the character of the user can be truly reflected by the music data, so that the insurance product can be recommended in a targeted manner, and the purchase conversion rate of the insurance product is improved; on the other hand, for the user who has the requirement of purchasing insurance, the insurance product meeting the requirement of the user can be timely obtained, the process is convenient, and the energy of the user is further saved.
In practical application, the terminal or the server is preset with an insurance product database, each insurance product in the insurance product database contains character labels of applicable groups, and the embodiment of the invention provides a method for selecting a target insurance product matched with the character labels from the preset insurance product database, as shown in fig. 2, and fig. 2 is a basic flow diagram of the method for selecting a target insurance product matched with the character labels from the preset insurance product database.
Specifically, as shown in fig. 2, step S1300 includes the steps of:
s1311, comparing the character label of the user with the label of the insurance product in the insurance product database;
the character labels of the users are labels corresponding to output values after feature cards of music activity data corresponding to the users are input into the character recognition neural network model. The tag user indicates the user's character classification, e.g., the tag may be marked as lively, powerful, perfect or flat, or as character features divided in other character divisions.
In the embodiment of the invention, each insurance product in the insurance database is marked with a label suitable for character characteristics, for example, a serious disease insurance product is marked with a perfect type, an X travel insurance is marked with an active type, and the insurance product is also marked with character labels divided by other character division modes. By comparing the user tag with text of the insurance product tag or a symbol of the user representing the tag feature.
S1312, when the comparison is consistent, taking the insurance product pointed by the label which is consistent in comparison as a target insurance product.
In the embodiment of the invention, if the comparison is consistent, the insurance product is determined to be matched with the user, and the insurance product pointed by the label is the target label insurance product. For example, when the label of the user and the label of the insurance product are both "perfect", the heavy-disease insurance product pointed by the "perfect" is determined as the target insurance product.
In practical applications, since the division of character features includes a plurality of kinds, and each user may have two kinds of character features, that is, in view of the middle of the two kinds of character features. In this case, when the user's character tag and the insurance product tag are multiple, the embodiment of the present invention provides a method for selecting the target insurance product matching the character tag from the preset insurance product database, as shown in fig. 3, fig. 3 is a basic flow diagram of a method for selecting the target insurance product matching the character tag from the preset insurance product database.
Specifically, as shown in fig. 3, step S1300 includes the steps of:
s1321, comparing the character label of the user with the label of the insurance product in the insurance product database;
in the embodiment of the present invention, the method for comparing the character tag of the user with the tag of the security product in the database of the security product refers to the embodiment shown in fig. 2, and will not be described herein.
S1322, when the character labels of the users and the labels of the insurance products in the insurance product database are the same, determining the number of the labels which are the same as the character labels of the users in each insurance product;
s1323, taking the insurance product with the largest label number as a target insurance product.
When the label comparison is consistent, in order to improve the matching accuracy, in the embodiment of the present invention, the number of consistent label comparison of the character label of the user and the insurance product is obtained, for example, the user has A, B, C and D four labels, the insurance product 1 has A, B and C three labels, the insurance product 2 has A, B, D and E four labels, the insurance product 3 has a and C two labels, and since the insurance product 1 has three identical labels with the user, the insurance product 2 and the insurance product 3 have two identical labels with the user, and thus, it is determined that the insurance product 1 is matched with the user, and the target insurance product of the user is the insurance product 1.
The embodiment of the invention also provides a training method of the personality neural network model, as shown in fig. 4, and fig. 4 is a basic flow diagram of the training method of the personality neural network model.
Specifically, as shown in fig. 4, the following steps are further included before step S1200:
s1210, acquiring music sample data marked with character labels;
the music sample data is a training sample image set for training a convolutional neural network model, and is an audio set comprising a plurality of character features, and the set is divided into a plurality of groups, wherein each group comprises a plurality of audios marked with the same character features.
The convolutional neural network model is a CNN convolutional neural network model or a VGG convolutional neural network model.
S1220, training a preset convolutional neural network model according to the music sample data to obtain a character recognition neural network model for recognizing character labels.
The embodiment of the invention also provides a method for training the preset convolutional neural network model according to the music sample data, as shown in fig. 5, fig. 5 is a basic flow diagram of the method for training the preset convolutional neural network model according to the music sample data.
Specifically, as shown in fig. 5, step S1220 includes the steps of:
s1221, extracting music characteristics from the music sample data, and calculating expected values of the music characteristics in each character label;
in the embodiment of the invention, the preset software can be utilized to extract the music characteristics from the music sample data, such as PC Sound Spectrum software, FFT spectrum analysis software, smartLive software and the like. In practical applications, in order to make frequencies in the frequency spectrum continuous and clear, pre-emphasis, windowing and fourier transform processing are usually performed on the audio to be evaluated in the process of converting the frequency spectrum.
In the embodiment of the invention, the audio file is extracted by softmax software.
And inputting the music characteristics of each group of character characteristics into a convolutional neural network model to obtain the output value of each music characteristic, sequencing the output values, and taking the intermediate value as the expected value of each group of character characteristics.
S1222, inputting the music characteristics into a preset convolutional neural network model to obtain excitation values of each character label in the convolutional neural network model;
in the embodiment of the invention, the music feature images of each group of character features are sequentially input into the neural network model, and the neural network model performs feature extraction and classification on the music feature images.
The excitation value is excitation data output by the convolutional neural network model according to the input music characteristic image, the excitation value is a numerical value with large discreteness before the neural network model is not trained to be converged, and the excitation value is relatively stable data after the neural network model is trained to be converged.
S1223, comparing whether the distance between the expected value and the excitation value of each character label is smaller than or equal to a preset threshold value, and when the distance between the expected value and the excitation value is larger than the threshold value, repeatedly and circularly iterating to update the weight in the convolutional neural network model through a reverse algorithm until the distance between the expected value and the excitation value is smaller than or equal to a preset first threshold value.
Judging whether the excitation value output by the neural network model full-connection layer is consistent with the set expected value or not through the loss function, and adjusting the weight in the first channel through a back propagation algorithm when the result is inconsistent.
In some embodiments, the loss function determines whether the excitation classification value is consistent with the set desired classification value by calculating a distance (euclidean distance or spatial distance) between the excitation classification value and the set desired classification value, sets a first threshold, determines that the excitation classification value is consistent with the set desired value when the distance between the excitation value and the set desired classification value is less than or equal to the first threshold, and otherwise, the excitation value is inconsistent with the set desired value.
When the excitation value of the neural network model is inconsistent with the set expected value, the weight in the neural network model needs to be corrected by adopting a random gradient descent algorithm, so that the output result of the convolutional neural network model is identical with the expected result of the classification judgment information. Through a plurality of training sample sets (in some embodiments, the photos in all training sample sets are disordered for training during training so as to increase the interference leaning capability of the model and enhance the stability of output, through repeated training and correction, when the comparison of the output classification data of the neural network model and the classification reference information of each training sample reaches (is not limited to) 99.5%, the training is ended.
In order to solve the technical problems, the embodiment of the invention also provides an insurance product recommending device. Referring specifically to fig. 6, fig. 6 is a basic structural block diagram of an insurance product recommendation device according to the present embodiment.
As shown in fig. 6, an insurance product recommendation device includes: an acquisition module 2100, a processing module 2200, and an execution module 2300. Wherein, the obtaining module 2100 is configured to obtain music activity data of a user; the processing module 2200 is configured to input the characteristics of the music activity data into a preset character recognition neural network model to obtain a character label of the user; and the execution module 2300 is used for selecting a target insurance product matched with the character label from a preset insurance product database and pushing the target insurance product to the webpage in which the user logs in.
The insurance product recommendation device obtains the character label of the user by inputting the characteristics of the music activity data of the user into a preset character recognition neural network model, and recommends the insurance product according to the character label of the user, and the characteristics of the user can be truly reflected by the music data, so that the insurance product can be pointedly recommended on one hand, and the purchase conversion rate of the insurance product is improved; on the other hand, for the user who has the requirement of purchasing insurance, the insurance product meeting the requirement of the user can be timely obtained, the process is convenient, and the energy of the user is further saved.
In some embodiments, each insurance product comprises a character tag of the applicable group in the insurance product database; the execution module comprises: the first processing sub-module is used for comparing the character label of the user with the label of the security product in the security product database; and the first execution sub-module is used for taking the insurance product pointed by the label which is consistent in comparison as the target insurance product when the comparison is consistent.
In some embodiments, when the user's character tag and the insurance product's tag are multiple, the execution module includes: the second processing sub-module is used for comparing the character label of the user with the label of the security product in the security product database; a second execution sub-module, configured to determine, when the user's character tag and the insurance product tag in the insurance product database have the same tag, the number of tags in each insurance product that is the same as the user's character tag; and the third execution sub-module is used for taking the insurance product with the largest label number as the target insurance product.
In some embodiments, further comprising: and the third processing sub-module is used for extracting the characteristics of the music activity data through preset characteristic extraction software.
In some embodiments, further comprising: the first acquisition submodule is used for acquiring music sample data marked with character labels; and the fourth processing sub-module is used for training a preset convolutional neural network model according to the music sample data to obtain a character recognition neural network model for recognizing the character label.
In some embodiments, the fourth processing sub-module comprises: the second acquisition submodule is used for extracting music characteristics from the music sample data and calculating expected values of the music characteristics in each character label; a fifth processing sub-module, configured to input the music feature into a preset convolutional neural network model, to obtain an excitation value of each character label in the convolutional neural network model; and the fourth execution sub-module is used for comparing whether the distance between the expected value and the excitation value of each character label is smaller than or equal to a preset threshold value, and updating the weight in the convolutional neural network model through a reverse algorithm in a repeated loop iteration mode when the distance between the expected value and the excitation value is larger than the threshold value, and ending when the distance between the expected value and the excitation value is smaller than or equal to a preset first threshold value.
In order to solve the technical problems, the embodiment of the invention also provides computer equipment. Referring specifically to fig. 7, fig. 7 is a basic structural block diagram of a computer device according to the present embodiment.
As shown in fig. 7, the internal structure of the computer device is schematically shown. As shown in fig. 7, the computer device includes a processor, a non-volatile storage medium, a memory, and a network interface connected by a system bus. The nonvolatile storage medium of the computer device stores an operating system, a database and computer readable instructions, the database can store a control information sequence, and the computer readable instructions can enable the processor to realize an insurance product recommendation method when the computer readable instructions are executed by the processor. The processor of the computer device is used to provide computing and control capabilities, supporting the operation of the entire computer device. The memory of the computer device may have stored therein computer readable instructions that, when executed by the processor, cause the processor to perform an insurance product recommendation method. The network interface of the computer device is for communicating with a terminal connection. It will be appreciated by those skilled in the art that the structure shown in fig. 7 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
The processor in this embodiment is configured to execute the specific contents of the acquisition module 2100, the processing module 2200, and the execution module 2300 in fig. 6, and the memory stores program codes and various types of data required for executing the above modules. The network interface is used for data transmission between the user terminal or the server. The memory in this embodiment stores program codes and data required for executing all the sub-modules in the insurance product recommendation method, and the server can call the program codes and data of the server to execute the functions of all the sub-modules.
The computer equipment obtains character labels of the users by inputting the characteristics of the music activity data of the users into a preset character recognition neural network model, and recommends insurance products according to the character labels of the users, and as the characteristics of the characters of the users can be truly reflected by the music data, the insurance products can be pointedly recommended on one hand, and the purchase conversion rate of the insurance products is improved; on the other hand, for the user who has the requirement of purchasing insurance, the insurance product meeting the requirement of the user can be timely obtained, the process is convenient, and the energy of the user is further saved.
The present invention also provides a storage medium storing computer readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the insurance product recommendation method of any of the embodiments described above.
Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored in a computer-readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. The storage medium may be a nonvolatile storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a random access Memory (Random Access Memory, RAM).
It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited in order and may be performed in other orders, unless explicitly stated herein. Moreover, at least some of the steps in the flowcharts of the figures may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, the order of their execution not necessarily being sequential, but may be performed in turn or alternately with other steps or at least a portion of the other steps or stages.
The foregoing is only a partial embodiment of the present invention, and it should be noted that it will be apparent to those skilled in the art that modifications and adaptations can be made without departing from the principles of the present invention, and such modifications and adaptations are intended to be comprehended within the scope of the present invention.
Claims (9)
1. A method of recommending insurance products, comprising the steps of:
acquiring music activity data of a user;
inputting the characteristics of the music activity data into a preset character recognition neural network model to obtain character labels of the users; wherein the musical activity data includes an audio spectrum; the character recognition neural network model is obtained by training a convolutional neural network based on music sample data marked with character labels;
selecting a target insurance product matched with the character label from a preset insurance product database, and pushing the target insurance product to a webpage in which the user logs in, wherein the method comprises the following steps: each insurance product is configured with at least one label associated with a character feature, the character label of the user is compared with the labels of the insurance products in the insurance product database, when the character label of the user is the same as the label of the insurance product in the insurance product database, the number of the labels which are the same as the character label of the user in each insurance product is determined, and the insurance product with the largest number of the labels is taken as the target insurance product.
2. The method of claim 1, wherein each insurance product comprises character tags of applicable groups in the insurance product database; the selecting the target insurance product matched with the character label from a preset insurance product database comprises the following steps:
comparing the character label of the user with the label of the security product in the security product database;
and when the comparison is consistent, taking the insurance product pointed by the label which is consistent in comparison as the target insurance product.
3. The method for recommending insurance products according to claim 1, wherein said inputting the characteristics of said musical activity data into a predetermined character recognition neural network model, before obtaining the character label of said user, further comprises
And extracting the characteristics of the music activity data through preset characteristic extraction software.
4. A recommendation method for an insurance product according to any one of claims 1 to 3, wherein before said inputting said music data into a predetermined character recognition neural network model to obtain a character label of said user, further comprising:
acquiring music sample data marked with character labels;
training a preset convolutional neural network model according to the music sample data to obtain a character recognition neural network model for recognizing the character label.
5. The method of claim 4, wherein training a predetermined convolutional neural network model based on the music sample data comprises:
extracting music characteristics from the music sample data, and calculating expected values of the music characteristics in each character label;
inputting the music characteristics into a preset convolutional neural network model to obtain excitation values of each character label in the convolutional neural network model;
comparing whether the distance between the expected value and the excitation value of each character label is smaller than or equal to a preset threshold value, and updating the weight in the convolutional neural network model through a reverse algorithm repeatedly and circularly when the distance between the expected value and the excitation value is larger than the threshold value, and ending when the distance between the expected value and the excitation value is smaller than or equal to a preset first threshold value.
6. An insurance product recommendation device, comprising:
the acquisition module is used for acquiring music activity data of the user;
the processing module is used for inputting the characteristics of the music activity data into a preset character recognition neural network model to obtain character labels of the users; wherein the musical activity data includes an audio spectrum; the character recognition neural network model is obtained by training a convolutional neural network based on music sample data marked with character labels;
the execution module is used for selecting a target insurance product matched with the character label from a preset insurance product database and pushing the target insurance product to the webpage in which the user logs in, and comprises the following steps: each insurance product is configured with at least one label associated with a character feature, the character label of the user is compared with the labels of the insurance products in the insurance product database, when the character label of the user is the same as the label of the insurance product in the insurance product database, the number of the labels which are the same as the character label of the user in each insurance product is determined, and the insurance product with the largest number of the labels is taken as the target insurance product.
7. The insurance product recommendation device of claim 6, wherein each insurance product in said insurance product database contains character labels of applicable groups; the execution module comprises:
the first processing sub-module is used for comparing the character label of the user with the label of the security product in the security product database;
and the first execution sub-module is used for taking the insurance product pointed by the label which is consistent in comparison as the target insurance product when the comparison is consistent.
8. A computer device comprising a memory and a processor, the memory having stored therein computer readable instructions that, when executed by the processor, cause the processor to perform the steps of the insurance product recommendation method of any of claims 1 to 5.
9. A storage medium storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the insurance product recommendation method of any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811347747.9A CN109360072B (en) | 2018-11-13 | 2018-11-13 | Insurance product recommendation method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811347747.9A CN109360072B (en) | 2018-11-13 | 2018-11-13 | Insurance product recommendation method and device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109360072A CN109360072A (en) | 2019-02-19 |
| CN109360072B true CN109360072B (en) | 2023-06-23 |
Family
ID=65344987
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811347747.9A Active CN109360072B (en) | 2018-11-13 | 2018-11-13 | Insurance product recommendation method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109360072B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110135942A (en) * | 2019-04-12 | 2019-08-16 | 深圳壹账通智能科技有限公司 | Products Show method, apparatus and computer readable storage medium |
| CN112991025A (en) * | 2021-05-08 | 2021-06-18 | 明品云(北京)数据科技有限公司 | Intelligent insurance recommendation method, system and equipment and computer readable storage medium |
| CN113256397B (en) * | 2021-07-02 | 2021-10-26 | 佛山市墨纳森智能科技有限公司 | Commodity recommendation method and system based on big data and computer-readable storage medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108492194A (en) * | 2018-03-06 | 2018-09-04 | 平安科技（深圳）有限公司 | Products Show method, apparatus and storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10628870B2 (en) * | 2016-10-10 | 2020-04-21 | International Business Machines Corporation | Offering personalized and interactive decision support based on learned model to predict preferences from traits |
| CN108108465A (en) * | 2017-12-29 | 2018-06-01 | 北京奇宝科技有限公司 | A kind of method and apparatus for pushing recommendation |
| CN108446688B (en) * | 2018-05-28 | 2020-01-07 | 北京达佳互联信息技术有限公司 | Face image gender judgment method and device, computer equipment and storage medium |
-
2018
- 2018-11-13 CN CN201811347747.9A patent/CN109360072B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108492194A (en) * | 2018-03-06 | 2018-09-04 | 平安科技（深圳）有限公司 | Products Show method, apparatus and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109360072A (en) | 2019-02-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11803747B2 (en) | Device placement optimization with reinforcement learning | |
| US20240078386A1 (en) | Methods and systems for language-agnostic machine learning in natural language processing using feature extraction | |
| CN108628971B (en) | Text classification method, text classifier and storage medium for unbalanced data set | |
| CN107832414A (en) | Method and apparatus for pushed information | |
| CN110020009B (en) | Online question and answer method, device and system | |
| CN107590224B (en) | Big data based user preference analysis method and device | |
| CN109360072B (en) | Insurance product recommendation method and device, computer equipment and storage medium | |
| CN107256267A (en) | Querying method and device | |
| US20120143907A1 (en) | Generating audio annotations for search and retrieval | |
| CN103425691A (en) | Search method and search system | |
| CN114065750A (en) | Commodity information matching and publishing method and device, equipment, medium and product thereof | |
| US10915586B2 (en) | Search engine for identifying analogies | |
| CN112632256A (en) | Information query method and device based on question-answering system, computer equipment and medium | |
| CN113722438A (en) | Sentence vector generation method and device based on sentence vector model and computer equipment | |
| CN110727871A (en) | Multi-mode data acquisition and comprehensive analysis platform based on convolution decomposition depth model | |
| CN111078849A (en) | Method and apparatus for outputting information | |
| CN113392920B (en) | Method, apparatus, device, medium, and program product for generating cheating prediction model | |
| CN114626926A (en) | Commodity search category identification method and device, equipment, medium and product thereof | |
| US20220414737A1 (en) | Query-based product representations | |
| CN113095723A (en) | Coupon recommendation method and device | |
| CN113792952A (en) | Method and apparatus for generating a model | |
| CN111026908B (en) | Song label determining method, device, computer equipment and storage medium | |
| KR102451020B1 (en) | A method of company-customized intelligent content curation using web crawling function | |
| CN115827864A (en) | Processing method for automatic classification of bulletins | |
| CN109241296A (en) | Method and apparatus for generating information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |