CN117217866A

CN117217866A - Medical commodity recommendation method and device, computer equipment and storage medium

Info

Publication number: CN117217866A
Application number: CN202311176289.8A
Authority: CN
Inventors: 吴汉
Original assignee: Kangjian Information Technology Shenzhen Co Ltd
Current assignee: Kangjian Information Technology Shenzhen Co Ltd
Priority date: 2023-09-12
Filing date: 2023-09-12
Publication date: 2023-12-12

Abstract

The application relates to the technical field of data processing, is suitable for the field of medical health, and discloses a medical commodity recommending method, a device, computer equipment and a storage medium, wherein the method comprises the following steps: responding to a commodity display request of a target user generated by triggering of a doctor end, and analyzing inquiry log information of the target user corresponding to the commodity display request; screening preset commodities according to the inquiry log information to determine a first commodity; screening the preset commodity according to the heat information of the preset commodity in the target period of the current moment to determine a second commodity; if the inquiry log information accords with the condition information of the target activity, determining an activity commodity corresponding to the target activity as a third commodity; performing duplicate removal processing on the first commodity, the second commodity and the third commodity to determine recommended commodity; and sending commodity information of the recommended commodity to a doctor. Therefore, the comprehensiveness and the safety of commodity recommendation are improved, the difficulty of selecting medical commodities by doctors is reduced, and the inquiry efficiency is improved.

Description

Medical commodity recommendation method and device, computer equipment and storage medium

Technical Field

The application relates to the technical field of data processing, and is suitable for the field of medical health, in particular to a medical commodity recommending method, a medical commodity recommending device, computer equipment and a storage medium.

Background

When on-line or off-line consultation is performed, a doctor needs to recommend commodities such as medicines or medical consumables to the patient according to the condition of the patient and the diagnosis result, and at this time, the doctor needs to search in-store commodities by experience and memory, and the non-satisfactory commodities are removed, so that the workload of the doctor is high, and the recommended commodities have high requirements on the skill level of the doctor.

In the related art, most of recommendation systems adopt a keyword matching search method to find commodities needed by doctors from a commodity library, are limited to user habit recommendation, and cannot actively recommend commodities suitable for patients according to personal information of the patients, current disease characteristics, welfare activities and other conditions, so that the conditions that the selected or opened commodities cannot be avoided and the conditions of the patients conflict, the safety and efficiency of commodity pushing are difficult to ensure, and commodity selection still completely depends on control of doctors, and cannot play a role in assisting doctors in medicine opening and recommending medicines.

Disclosure of Invention

In view of the above, the application provides a medical commodity recommending method, a medical commodity recommending device, a medical commodity recommending computer device and a medical commodity storing medium, so as to solve the problems that medical commodities are difficult to screen by doctors and the inquiry efficiency is low.

In a first aspect, a method for recommending medical goods is provided, including:

responding to a commodity display request of a target user generated by triggering of a doctor end, and analyzing inquiry log information of the target user corresponding to the commodity display request;

screening preset commodities according to the inquiry log information to determine a first commodity;

screening the preset commodity according to the heat information of the preset commodity in the target period of the current moment to determine a second commodity;

if the inquiry log information accords with the condition information of the target activity, determining an activity commodity corresponding to the target activity as a third commodity;

performing duplicate removal processing on the first commodity, the second commodity and the third commodity to determine recommended commodity;

and sending the commodity information of the recommended commodity to a doctor end so that the doctor end displays the commodity information of the recommended commodity.

Further, screening the preset commodity according to the inquiry log information to determine a first commodity, including:

performing feature extraction processing on the inquiry log information to determine inquiry features of the inquiry log information;

inputting the inquiry features into a recommendation model to obtain a first commodity and the matching degree of the first commodity and the inquiry features;

and if the number of the first commodities is larger than the preset number, filtering the first commodities according to the preset number and the matching degree.

Further, before inputting the inquiry feature into the recommendation model, the recommendation method of the medical commodity further comprises:

acquiring historical recommended commodities aiming at different target users and historical order information of different target users sent by a doctor side;

determining the historical recommended commodity and the order commodity in the historical order information as preset commodity;

determining historical inquiry features corresponding to the historical recommended commodities and historical inquiry features corresponding to the historical order information as inquiry feature labels of preset commodities;

training the preset model according to the preset commodity and the inquiry feature label to obtain a recommended model.

Further, screening the preset commodity according to the heat information of the preset commodity in the target period to which the current moment belongs, including:

determining a preset commodity with the heat information conforming to the preset heat information as a second commodity;

the heat information comprises recommended times of preset commodities, searching times of the preset commodities, sales amount of the preset commodities and/or evaluation amount of the preset commodities.

Further, the medical commodity recommending method further comprises the following steps:

determining a preset period to which the current moment belongs;

and calculating a target time period corresponding to the history time of the same ratio or the ring ratio according to the time minimum value and the time maximum value of the preset time period.

if the user characteristics corresponding to the inquiry log information do not accord with the applicable condition information of the recommended commodity, deleting the recommended commodity.

acquiring positioning information of a target user;

determining a target merchant within a distribution position range corresponding to the positioning information;

if the stock quantity of the recommended commodity corresponding to the target merchant does not accord with the demand quantity of the recommended commodity, deleting the recommended commodity;

if the stock quantity of the recommended commodity corresponding to the target merchant accords with the demand quantity of the recommended commodity, the navigation information of the target merchant is sent to the user side of the target user in response to the order confirmation instruction of the recommended commodity.

sorting the recommended commodities according to a plurality of preset sorting strategies to obtain a plurality of recommendation lists of the recommended commodities;

and sending the plurality of recommendation lists to a doctor terminal so that the doctor terminal responds to the display instruction of the target recommendation list and displays commodity information of the recommended commodities according to the sequence of the target recommendation list.

Further, before the recommended commodities are ranked according to the plurality of preset ranking strategies, the medical commodity recommending method further comprises the following steps:

Under the condition that a preset ordering strategy is arranged according to the weight scores, determining the weight scores of the first commodity in the recommended commodity according to the matching degree of the first commodity and the inquiry feature and the first weight;

determining a weight score of a second commodity in the recommended commodity according to the heat information and the second weight;

and determining the weight score of the third commodity in the recommended commodity according to the priority of the target activity and the third weight.

In a second aspect, there is provided a medical commodity recommending apparatus, comprising:

the analysis module is used for responding to the commodity display request of the target user generated by the triggering of the doctor end and analyzing the inquiry log information of the target user corresponding to the commodity display request;

the screening module is used for screening preset commodities according to the inquiry log information to determine a first commodity; screening the preset commodity according to the heat information of the preset commodity in the target period of the current moment to determine a second commodity; if the inquiry log information accords with the condition information of the target activity, determining an activity commodity corresponding to the target activity as a third commodity; performing duplicate removal processing on the first commodity, the second commodity and the third commodity to determine recommended commodity;

And the communication module is used for sending the commodity information of the recommended commodity to the doctor end so that the doctor end can display the commodity information of the recommended commodity.

Further, the screening module is specifically configured to perform feature extraction processing on the inquiry log information, and determine inquiry features of the inquiry log information; inputting the inquiry features into a recommendation model to obtain a first commodity and the matching degree of the first commodity and the inquiry features; and if the number of the first commodities is larger than the preset number, filtering the first commodities according to the preset number and the matching degree.

Further, the medical commodity recommending apparatus further includes:

the acquisition module is used for acquiring historical recommended commodities for different target users and historical order information of the different target users sent by the doctor side;

the sample determining module is used for determining the historical recommended commodities and the ordered commodities in the historical order information as preset commodities; and determining the historical inquiry characteristics corresponding to the historical recommended commodity and the historical inquiry characteristics corresponding to the historical order information as inquiry characteristic labels of preset commodities;

the training module is used for training the preset model according to the preset commodity and the inquiry feature label to obtain a recommended model.

Further, the screening module is specifically configured to determine a preset commodity with heat information according to the preset heat information as a second commodity; the heat information comprises recommended times of preset commodities, searching times of the preset commodities, sales amount of the preset commodities and/or evaluation amount of the preset commodities.

Further, the medical commodity recommending apparatus further includes:

the time determining module is used for determining a preset time period to which the current moment belongs; and calculating a target time period corresponding to the history time of the same ratio or the ring ratio according to the time minimum value and the time maximum value of the preset time period.

Further, the screening module is further configured to delete the recommended commodity if the user characteristic corresponding to the inquiry log information does not conform to the applicable condition information of the recommended commodity.

Further, the medical commodity recommending apparatus further includes:

the position determining module is used for acquiring positioning information of the target user;

the screening module is also used for determining a target merchant in the distribution position range corresponding to the positioning information; if the stock quantity of the recommended commodity corresponding to the target merchant does not accord with the demand quantity of the recommended commodity, deleting the recommended commodity;

and the communication module is also used for responding to the order confirmation instruction of the recommended commodity and sending the navigation information of the target merchant to the user side of the target user if the stock quantity of the recommended commodity corresponding to the target merchant meets the demand quantity of the recommended commodity.

Further, the medical commodity recommending apparatus further includes:

the ordering module is used for ordering the recommended commodities according to a plurality of preset ordering strategies to obtain a plurality of recommendation lists of the recommended commodities;

and the communication module is also used for sending the plurality of recommendation lists to the doctor end so that the doctor end responds to the display instruction of the target recommendation list and displays commodity information of the recommended commodities according to the sequence of the target recommendation list.

Further, the medical commodity recommending apparatus further includes:

the scoring module is used for determining the weight score of the first commodity in the recommended commodity according to the matching degree of the first commodity and the inquiry feature and the first weight under the condition that the preset ordering strategy is arranged according to the weight score; determining a weight score of a second commodity in the recommended commodity according to the heat information and the second weight; and determining the weight score of the third commodity in the recommended commodity according to the priority of the target activity and the third weight.

In a third aspect, a computer device is provided comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor executing the computer program to perform the steps of the method of recommending medical items as described above.

In a fourth aspect, a computer readable storage medium is provided, the computer readable storage medium storing a computer program which, when executed by a processor, implements the steps of the medical commodity recommendation method described above.

In the proposal realized by the medical commodity recommending method, the medical commodity recommending device, the computer equipment and the storage medium, when a doctor needs to search the commodity suitable for the target user, the doctor can report the inquiry log information of the target user to the server through the doctor terminal. And the server responds to the commodity display request of the target user generated by the triggering of the doctor terminal, and analyzes inquiry log information of the target user carried by the commodity display request. And screening out first commodities which accord with the disease types and symptom characteristics corresponding to the diseases of the target user according to the inquiry log information. Meanwhile, the second commodity which accords with the fashion trend can be screened out by utilizing the heat information in the target period, and the movable commodity (third commodity) which is suitable for the target user can be screened out. And carrying out de-duplication treatment on the first commodity, the second commodity and the third commodity to obtain the multi-dimensional recommended commodity with comprehensive heat tendency, user disease condition and special activities. And finally, sending commodity information of the recommended commodity to a doctor end for display. On the one hand, according to the user information and the disease condition of the target user, the commodity suitable for the target user is recommended to the doctor, the difficulty of selecting medical commodity by the doctor is reduced, the commodity recommendation safety is ensured, and the efficiency of screening medical commodity by the doctor is improved. On the other hand, on the basis of considering the basic inquiry condition of the user, the heat trend and marketing activity strategy of the commodities in the warehouse are comprehensively considered, so that the commodity recommendation breaks through the limitation of habit recommendation of a single object, the comprehensiveness of commodity recommendation is improved, the effectiveness and timeliness of medical commodity information pushing are improved, the recommended commodities can quickly respond to the current actual life condition, a doctor can quickly find out the commodities meeting the inquiry requirements of a target user through the commodities recommended by the system, or fine adjustment is performed on the basis of recommending the commodities by the system, and the workload of the doctor is greatly saved.

The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments of the present application will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, 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 schematic view of an application environment of a method of recommending medical goods in the present application;

FIG. 2 is a flow chart of a method of recommending medical goods in accordance with the present application;

FIG. 3 is a schematic structural view of a medical commodity recommending apparatus according to the present application;

fig. 4 is a schematic structural diagram of a computer device in the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

The medical commodity recommending method provided by the application can be applied to an application environment shown in figure 1. Wherein, doctor end 101 communicates with server 102 via a network. The doctor terminal 101 may be, but not limited to, various personal computers, notebook computers, smartphones, tablet computers and portable wearable devices, and the server 102 may be implemented by a stand-alone server or a server cluster formed by a plurality of servers. The server 102 can screen out recommended goods in response to the goods presentation request of the target user generated by the triggering of the doctor terminal 101, and feed back the recommended goods to the doctor terminal 101.

The server 102 of embodiments of the present application may acquire and process relevant data based on artificial intelligence techniques. Among these, artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use knowledge to obtain optimal results. Artificial intelligence infrastructure technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like.

Referring to fig. 2, the present embodiment provides a medical commodity recommendation method, which is described by taking the application of the method to the server 102 in fig. 1 as an example, and includes the following steps:

s10, responding to a commodity display request of a target user generated by the triggering of a doctor end, and analyzing inquiry log information of the target user corresponding to the commodity display request.

The target user refers to a patient to be diagnosed or an agent of the patient. The inquiry log information of the target user refers to inquiry contents describing personal information of the user and related disorders thereof, such as surname, age, usual location, occupation, smoking, stay up, allergy history, pregnancy, etc. data describing characteristics of the user, and data describing characteristics of the disease, such as medical history of the user, doctor's diagnosis result, etc. The embodiment of the present application is not limited thereto.

It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and the data (including but not limited to data for analysis, stored data, presented data, etc.) related to the present application are information and data authorized by the user or fully authorized by each party, and the collection, use and processing of the related data need to comply with the related laws and regulations and standards of the related country and region, and provide corresponding operation entries for the user to select authorization or rejection. And declaring a data privacy protocol to keep user information and inquiry log information secret.

S20, screening preset commodities according to the inquiry log information, and determining a first commodity.

The preset commodities comprise medical commodities which are searched or ordered by doctors, so that the preset commodities and corresponding inquiry information thereof are stored in the system, and screening of the preset commodities through inquiry log information is conveniently achieved. The related information of the preset commodity in the application can be from an authoritative website or related database of an international disease classification system, a national food and drug administration, a United states food and drug administration and the like, and from a hospital, a physical pharmacy, an online pharmacy and the like.

In this embodiment, when the doctor needs to find out the commodity suitable for the target user, the doctor can report the inquiry log information of the target user to the server. And the server responds to the commodity display request of the target user generated by the triggering of the doctor terminal, and analyzes inquiry log information of the target user carried by the commodity display request. And screening out first commodities which accord with the disease types and symptom characteristics corresponding to the diseases of the target user according to the inquiry log information. Therefore, the system server recommends the commodity suitable for the target user to the doctor according to the user information and the condition of the target user, and the doctor can quickly find out the commodity meeting the inquiry requirement of the target user through the commodity recommended by the system, so that the difficulty of selecting medical commodity by the doctor is reduced, the safety of commodity recommendation is ensured, and the workload of the doctor is greatly saved.

In a specific application scenario, step S20, namely screening preset commodities according to inquiry log information, determines a first commodity, specifically includes the following steps:

s21, carrying out feature extraction processing on the inquiry log information, and determining inquiry features of the inquiry log information.

Wherein the inquiry features include user features and disease features.

In the embodiment, through the feature extraction processing, the key features which are simpler and have higher relevance to the inquiry situation of the target user can be extracted from the original inquiry log information, so that the redundancy and complexity of data are reduced, the subsequent data analysis can be focused on important information, the efficiency of data processing and analysis is improved, and the computing resources and the time cost are saved. And the inquiry log information can be converted into structured data, so that various data analysis, storage and mining can be conveniently performed.

S22, inputting the inquiry features into a recommendation model to obtain a first commodity and the matching degree of the first commodity and the inquiry features.

Specifically, the index value of each element in the input vector can be calculated through a softmax function configured in the recommendation model, all the index values are added to obtain a sum, each index value is divided by the sum to obtain a normalized probability value, and the normalized probability vector is returned to serve as the matching degree output of the first commodity and the inquiry feature. Wherein the softmax function is defined as follows: Where z represents the j-th element of the input vector and K represents the length of the vector.

It should be noted that, before step S22, the recommendation model needs to be trained by preset merchandise. Specifically, historical recommended commodities for different target users and historical order information of different target users sent by a doctor side are obtained; determining the historical recommended commodity and the order commodity in the historical order information as preset commodity; determining historical inquiry features corresponding to the historical recommended commodities and historical inquiry features corresponding to the historical order information as inquiry feature labels of preset commodities; training the preset model according to the preset commodity and the inquiry feature label to obtain a recommended model.

The preset model may be a pre-trained single neural Network model, such as a deep neural Network (Deep Neural Networks, DNN) model, a convolutional Network (Convolutional Neural Networks, CNN) model, a cyclic neural Network (Recurrent Neural Networks, RNN) model, a Residual Network (Residual Network) model, a BERT (Bidirectional Encoder Representation from Transformers, deep bi-directional self-attention Network), a robert (Robustly Optimized BERT Pretraining Approach, robust deep bi-directional self-attention Network) or a MacBERT (MLM as correction BERT, mask corrected deep bi-directional self-attention Network), or a multi-Network parallel dual Network structure model (connector), such as a connector model composed of CNN and a converter branch, wherein the CNN branch adopts a Residual Network structure and the converter branch adopts a ViT structure.

In this embodiment, the server extracts from the database the historical recommended goods for the different target users, i.e. the recommended goods once sent to the doctor's end, and the historical order information of the different target users once ordered by the doctor. The commodity screening result and the doctor selection result of the system during inquiry of different users can be determined through the historical recommended commodities and the order commodities in the historical order information, and the historical recommended commodities and the order commodities in the historical order information are determined to be preset commodities which can be used as model training samples. Meanwhile, the server can search the historical inquiry features corresponding to the historical recommended commodities and the historical order information from the database, and take the historical inquiry features corresponding to the historical recommended commodities and the historical order information as inquiry feature labels of preset commodities. And carrying out repeated iterative training on the preset model by utilizing preset commodity and inquiry feature labels, and updating and optimizing the model in each iteration to finally obtain a recommended model. So that the screening of the first commodity is conveniently carried out by using the recommendation model, the screening accuracy can be ensured, the stability and the repeatability of data processing are improved, the error and the uncertainty are reduced, and the labor and time cost are saved. And moreover, the automatic screening under the high concurrency scene is realized, so that more diversified data selection is realized, and different application scenes and requirements are met.

S23, if the number of the first commodities is larger than the preset number, filtering the first commodities according to the preset number and the matching degree.

In this embodiment, the first commodities are screened out by using a recommendation model trained in advance by preset commodities, and the matching degree of each first commodity and the inquiry feature of the target user is obtained. The higher the matching degree is, the more the first commodity meets the requirement of the target user for inquiry. Helping doctors to find out commodities suitable for target users more quickly, and reducing selection cost. Further, when the number of first merchandise is greater than the preset number, in order to avoid a large amount of useless information interfering with the doctor's decision. And screening out a preset number of first commodities with higher matching degree based on the matching degree output by the recommendation model. Therefore, the number of recommended commodities provided for doctors by the system is simplified while the recommended commodities meet the demands of target users, more accurate and personalized recommended results are provided by reducing the commodity selection range, the doctors are helped to quickly find the commodities meeting the demands of the target users, and more refined services are provided for the doctors and the patients.

S30, screening the preset commodity according to the heat information of the preset commodity in the target period to which the current moment belongs, and determining a second commodity.

In this embodiment, the preset commodity is screened by the heat information of the preset commodity in the target period to which the current time belongs. Therefore, the popular trend analysis is carried out on different preset commodities, second commodities with the rising trend of recent heat are mined, the second commodities are frequently inquired or used in the recent period, the commodities are recommended to doctors, the doctors can be helped to know recent hot spot symptoms and countermeasures, the explicit display of commodity trends is realized, convenience is brought to the doctors to conveniently select medical commodities required by the recent high-frequency diseases, and doctor inquiry efficiency is improved.

The method comprises the steps of acquiring corresponding commodity heat information according to a target period to which the current moment belongs, so that timeliness of the heat information is guaranteed, and a screened second commodity can be more fit with recent popular conditions. The target time period is used for limiting the acquisition time period of the heat information, can be one month or one quarter, and can be reasonably set according to the timeliness of commodity recommendation. The same or different target period may be set for different regions, for example, one year before the current time is taken as a target period for the medicine, thereby enabling the update of the heat information by the year; the medical consumable takes one month before the current moment as a target period, so that the heat information can be updated monthly.

It is noted that, before S20, the method for recommending medical goods further includes: determining a preset period to which the current moment belongs; and calculating a target time period corresponding to the history time of the same ratio or the ring ratio according to the time minimum value and the time maximum value of the preset time period.

In this embodiment, the server determines the preset period to which the current time belongs according to the division of the preset period. And performing the same-ratio or ring-ratio time dimension prediction on the preset time period to which the current moment belongs. Therefore, the preset commodity heat information counted in the predicted target period can be more in line with the actual situation in the counted period, so that seasonal difference can be eliminated, the fitting degree of the recommended commodity and the recent target user demand is improved, the medical commodity screening accuracy is further guaranteed, and the platform service efficiency is improved.

For example, when the period to which the current time belongs is 2014-10-11-2014-12-11, the target period with the same ratio is 2013-10-11-2013-12-11, and the target period with the ring ratio is 2014-08-11-2014-10-11.

In a specific application scenario, step S30, namely screening the preset commodity according to the heat information of the preset commodity in the target period to which the current time belongs, specifically includes the following steps:

S31, determining the preset commodity with the heat information conforming to the preset heat information as a second commodity.

In this embodiment, the heat information of the preset merchandise is compared with the preset heat information. If the heat information of the preset commodity accords with the preset heat information, the fact that the recent heat of the preset commodity has an ascending trend is indicated, and the preset commodity is frequently searched or ordered by doctors and possibly suitable for recent epidemic diseases. At this time, the preset commodity is determined as a recommendable second commodity. So that doctors can conveniently select medical commodities required by recent high-incidence diseases, and the doctor's inquiry efficiency for epidemic diseases is effectively improved.

For example, the recent (one week to one month) different patient diagnoses and merchandise sales are drawn from the library and the recommended frequency for each merchandise is determined. After the doctor reports the user information label of the current patient to be asked, the server screens high-frequency commodities (recently marketable commodities) suitable for the patient according to the label so as to facilitate the selection of the doctor.

And S40, if the inquiry log information accords with the condition information of the target activity, determining an activity commodity corresponding to the target activity as a third commodity.

The target activities can be issued by the medical institutions at irregular intervals by means of the server platform, and the server can crawl relevant information of the medical institutions from a plurality of medical institution platforms in a crawler mode. The target activity can be a popularization activity of new and old commodities, such as XX medicines newly marketed, XX medical consumables newly put in storage and the like; the target campaign may also be a promotional campaign for the good, e.g., drug a full 5 boxes giving 5 boxes, consume full 500 yuan giving 2 boxes of drug B, etc. The embodiment of the application is not particularly limited, and can be reasonably arranged according to the actual medical activity of the platform.

In this embodiment, the server judges whether the target activity is suitable for the target user by comparing the disease feature and/or the user feature extracted from the inquiry log information with the activity condition information of the target activity, and takes the activity commodity indicated by the target activity as a recommended third commodity if suitable. Therefore, the commodity database is perfected through movable commodities, the problem that the non-recommended or selected commodities cannot be hit by inquiry features and heat information can be avoided, the limitation of habit-shaped recommendation of a single object is broken through in commodity recommendation, the diversity of recommended commodities is enriched, and the efficiency of a doctor in finding out required commodities is improved. Meanwhile, the third commodity is pushed, so that the prompt effect on welfare commodity can be achieved, the cost of purchasing commodity by the target user is reduced, the waste of medical insurance resources is reduced, and the good sensitivity of the target user to doctors is improved in a better auxiliary mode.

For example, in a case where a patient is female and a doctor is in a consultation scenario for treating a disease a, the server pushes the drug C to a doctor as a recommended drug to prompt the doctor for a new product.

S50, performing duplicate removal processing on the first commodity, the second commodity and the third commodity, and determining recommended commodity.

In the embodiment, redundant data in the first commodity, the second commodity and the third commodity which are screened out through different dimensions are deleted through the de-duplication process, so that the recommended commodity only contains a unique and non-repeated object or record, and the consistency and the integrity of the recommended commodity are maintained. And the repeated data can be effectively prevented from occupying extra storage space of a doctor end or a server, the communication overhead between the server and the doctor end is reduced, and the operation cost of the platform is reduced.

Further, for better optimization of recommended merchandise, the recommended merchandise may be further filtered in the following manner. Specifically:

in the first mode, if the user characteristics corresponding to the inquiry log information do not accord with the applicable condition information of the recommended commodity, the recommended commodity is deleted.

In the embodiment, the contraindication relationship between the commodity and the user and the influence of the medicine on different target users are comprehensively considered, medicines which have risks on the target users in the recommended medicines are eliminated through the application conditions of the recommended commodity, and the accuracy and the safety of generating commodity recommendation information are improved.

For example, for male patients, gynecological medicines and female-dedicated instruments in recommended goods are deleted. Or for pregnant women, the system needs to delete drugs contraindicated for pregnant women.

Obtaining positioning information of a target user; determining a target merchant within a distribution position range corresponding to the positioning information; if the stock quantity of the recommended commodity corresponding to the target merchant does not accord with the demand quantity of the recommended commodity, deleting the recommended commodity; if the stock quantity of the recommended commodity corresponding to the target merchant accords with the demand quantity of the recommended commodity, the navigation information of the target merchant is sent to the user side of the target user in response to the order confirmation instruction of the recommended commodity.

The target merchant may be a target with medical commodity sales permission such as a hospital or a pharmacy.

In this embodiment, in the scenario of O2O (Online to Offline), the server can locate the target merchant within the range of the satisfied delivery location, i.e., capable of achieving the sale and delivery of the merchandise, by the location where the target user is located. When the stock quantity of the recommended commodity corresponding to the target merchant does not meet the demand quantity of the recommended commodity, the target merchant is not capable of providing enough commodity for the target user to purchase, and at the moment, the recommended commodity is deleted, so that doctors are prevented from recommending medical commodity which cannot be purchased to the target user, reliability of commodity recommendation is guaranteed, and doctors are assisted in selecting more suitable medical commodity for the target user. On the contrary, under the condition that the user side of the target user and the server establish a communication relationship, if the stock quantity of the recommended commodity corresponding to the target merchant accords with the demand quantity of the recommended commodity, namely, the target user can purchase the recommended commodity at the target merchant, and the server receives an order confirmation instruction of the target user for the recommended commodity, the server sends the navigation information of the target merchant to the user side of the target user, so that the target user can reach the sales place of the recommended commodity more quickly through the navigation information, and the use experience of the user is improved.

And S60, sending commodity information of the recommended commodity to a doctor terminal.

The preset commodity, the movable commodity and the recommended commodity in the application can be medical commodity such as medicines, medical consumables or medical services (such as physical examination packages and accessory treatment courses).

According to the medical commodity recommending method provided by the embodiment, when a doctor needs to search for a commodity suitable for a target user, the doctor can report inquiry log information of the target user to a server through the doctor terminal. And the server responds to the commodity display request of the target user generated by the triggering of the doctor terminal, and analyzes inquiry log information of the target user carried by the commodity display request. And screening out first commodities which accord with the disease types and symptom characteristics corresponding to the diseases of the target user according to the inquiry log information. Meanwhile, the second commodity which accords with the fashion trend can be screened out by utilizing the heat information in the target period, and the movable commodity (third commodity) which is suitable for the target user can be screened out. And carrying out de-duplication treatment on the first commodity, the second commodity and the third commodity to obtain the multi-dimensional recommended commodity with comprehensive heat tendency, user disease condition and special activities. And finally, sending commodity information of the recommended commodity to a doctor end for display. On the one hand, according to the user information and the disease condition of the target user, the commodity suitable for the target user is recommended to the doctor, the difficulty of selecting medical commodity by the doctor is reduced, the commodity recommendation safety is ensured, and the efficiency of screening medical commodity by the doctor is improved. On the other hand, on the basis of considering the basic inquiry condition of the user, the heat trend and marketing activity strategy of the commodities in the warehouse are comprehensively considered, so that the commodity recommendation breaks through the limitation of habit recommendation of a single object, the comprehensiveness of commodity recommendation is improved, the effectiveness and timeliness of medical commodity information pushing are improved, the recommended commodities can quickly respond to the current actual life condition, a doctor can quickly find out the commodities meeting the inquiry requirements of a target user through the commodities recommended by the system, or fine adjustment is performed on the basis of recommending the commodities by the system, and the workload of the doctor is greatly saved.

In some embodiments of the present application, there is provided a specific entity alignment scheme, before or after step S60, the medical commodity recommendation method further includes: sorting the recommended commodities according to a plurality of preset sorting strategies to obtain a plurality of recommendation lists of the recommended commodities; the plurality of recommendation lists are sent to the doctor side.

The preset sorting policy may be set reasonably according to the needs of a doctor or a user, for example, sorting according to stock quantity, sorting according to price, sorting according to recommended times, sorting according to matching degree, sorting according to commodity types, and the like.

In this embodiment, the recommended goods are respectively ranked by a plurality of different preset ranking strategies, so as to obtain a recommendation list of the same recommended goods based on a plurality of different display orders. After the recommendation lists are sent to the doctor side, the doctor can select the display sequence of the requirements through the doctor side. Therefore, the doctor terminal can display commodity information of the recommended commodities in sequence according to the target recommendation list selected by the doctor. Thereby providing a diversified information angle for doctors, helping the doctors to find required commodities more easily, and making more accurate and intelligent decisions so as to meet the personalized requirements of different target users.

Further, before the recommended commodities are ranked according to the plurality of preset ranking strategies, the medical commodity recommending method further comprises the following steps: under the condition that a preset ordering strategy is arranged according to the weight scores, determining the weight scores of the first commodity in the recommended commodity according to the matching degree of the first commodity and the inquiry feature and the first weight; determining a weight score of a second commodity in the recommended commodity according to the heat information and the second weight; and determining the weight score of the third commodity in the recommended commodity according to the priority of the target activity and the third weight.

In this embodiment, under the condition that the preset ranking policy is ranked according to the weight score, the matching degree of the first commodity and the inquiry feature, the heat information and the priority of the target activity are respectively and uniformly quantized through the weight of multiple dimensions, so that the recommended commodities screened out in different dimensions are ranked through normalization processing.

In one particular embodiment, a commodity recommendation method for a doctor to serve a patient in a session based on a consultation is provided.

(1) Extracting inquiry log records: the inquiry log record includes: user characteristics (including age, sex), department of diagnosis, and the like, and disease-related characteristics are extracted from information such as complaint content, inquiry dialogue content, diagnosis results of doctors, and the like. In particular, principal Component Analysis (PCA) methods may be utilized to find the principal features in the dataset.

(2) Constructing a commodity library: the commodity library comprises historical commodity transaction records, the occurrence frequency of the commodity and commodity applicable conditions.

(3) Model training: and training an AI model based on the characteristics of the commodities in the commodity library and the corresponding inquiry logs thereof so as to predict recommended commodities aiming at the target inquiry ticket.

(4) And predicting a first commodity list suitable for the questionnaire by using a trained artificial intelligent convolutional neural network commodity prediction model (wherein the first commodity in the list is medical commodity screened according to the age, sex and other conditions of a patient in the questionnaire and the characteristics of a doctor ID, a questionnaire dialogue between the doctor and the patient and the like, and then scoring and sorting the screened first commodity according to the matching condition of the commodity and the questionnaire to obtain the first commodity list). And extracting a specific number of first commodity outputs in the first commodity list. The matching of the merchandise to the questionnaire, i.e., the correlation, can be calculated from the softmax function.

(5) And (3) extracting the patient diagnosis and commodity pushing conditions in the near term (from one week to one month) from a commodity library to construct a frequency table, inputting the characteristic label of the current inquiry patient each time, and screening high-frequency commodities (near term marketable commodities) according to the label.

(6) And operating and configuring recent activity commodities, wherein the content of the configured activity commodities comprises commodity/service package information, applicable patient types (inquiry feature labels) and activity priorities (high, medium and low), and the frequency of mapping according to the activity priorities (the highest frequency of the high-priority mapping (5) and the commodities under the labels is the same as the highest frequency of the commodities). A table of frequency of the movable commodity is obtained.

(7) And (3) data filtering: and (5) carrying out de-duplication combination (the repeated part keeps the corresponding attribute) on the commodity list obtained in the steps (4), (5) and (6). And then carrying out safe filtration according to the commodity library and the acquired basic information of the patient to remove inapplicable commodities. And (3) carrying out inventory inquiry (inquiring the inventory and distribution condition of the commodities in stores/warehouses in five kilometers of the patient) according to the recommended commodities after screening and the current geographical position of the user, and filtering out the inquired commodities which do not support distribution or lack inventory. And finally, carrying out normalization processing according to the frequency of the recommended commodity, calculating the score by adopting a MIN-MAX algorithm, sorting the score of the result of the whole recommended commodity, and outputting the result to a doctor.

In the embodiment, the intelligent recommendation candidate can be provided for a doctor to push out commodities according to the model or according to various marketing requirements, the recommended commodities meet all basic requirements, a great deal of work can be saved for the doctor, and fine adjustment can be performed after the recommended commodities are checked.

It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic, and should not limit the implementation process of the embodiment of the present application.

In an embodiment, a medical commodity recommending device is provided, and the medical commodity recommending device corresponds to the medical commodity recommending method in the embodiment one by one. As shown in fig. 3, the medical commodity recommending apparatus includes an analyzing module 301, a screening module 302, and a communication module 303. The functional modules are described in detail as follows:

the analysis module 301 is configured to respond to a commodity display request of a target user generated by triggering of a doctor end, and analyze inquiry log information of the target user corresponding to the commodity display request;

the screening module 302 is configured to screen preset commodities according to the inquiry log information, and determine a first commodity; screening the preset commodity according to the heat information of the preset commodity in the target period of the current moment to determine a second commodity; if the inquiry log information accords with the condition information of the target activity, determining an activity commodity corresponding to the target activity as a third commodity; performing duplicate removal processing on the first commodity, the second commodity and the third commodity to determine recommended commodity;

the communication module 303 is configured to send the commodity information of the recommended commodity to the doctor end, so that the doctor end displays the commodity information of the recommended commodity.

The application provides a medical commodity recommending device, which can report inquiry log information of a target user to a server through a doctor terminal when the doctor needs to search commodities suitable for the target user. And the server responds to the commodity display request of the target user generated by the triggering of the doctor terminal, and analyzes inquiry log information of the target user carried by the commodity display request. And screening out first commodities which accord with the disease types and symptom characteristics corresponding to the diseases of the target user according to the inquiry log information. Meanwhile, the second commodity which accords with the fashion trend can be screened out by utilizing the heat information in the target period, and the movable commodity (third commodity) which is suitable for the target user can be screened out. And carrying out de-duplication treatment on the first commodity, the second commodity and the third commodity to obtain the multi-dimensional recommended commodity with comprehensive heat tendency, user disease condition and special activities. And finally, sending commodity information of the recommended commodity to a doctor end for display. On the one hand, according to the user information and the disease condition of the target user, the commodity suitable for the target user is recommended to the doctor, the difficulty of selecting medical commodity by the doctor is reduced, the commodity recommendation safety is ensured, and the efficiency of screening medical commodity by the doctor is improved. On the other hand, on the basis of considering the basic inquiry condition of the user, the heat trend and marketing activity strategy of the commodities in the warehouse are comprehensively considered, so that the commodity recommendation breaks through the limitation of habit recommendation of a single object, the comprehensiveness of commodity recommendation is improved, the effectiveness and timeliness of medical commodity information pushing are improved, the recommended commodities can quickly respond to the current actual life condition, a doctor can quickly find out the commodities meeting the inquiry requirements of a target user through the commodities recommended by the system, or fine adjustment is performed on the basis of recommending the commodities by the system, and the workload of the doctor is greatly saved.

In one embodiment, the screening module 302 is specifically configured to perform feature extraction processing on the inquiry log information, and determine inquiry features of the inquiry log information; inputting the inquiry features into a recommendation model to obtain a first commodity and the matching degree of the first commodity and the inquiry features; and if the number of the first commodities is larger than the preset number, filtering the first commodities according to the preset number and the matching degree.

In an embodiment, the medical commodity recommending device further includes: the acquisition module (not shown in the figure) is used for acquiring historical recommended commodities for different target users and historical order information of different target users sent by a doctor side; the sample determining module (not shown in the figure) is used for determining the historical recommended goods and the order goods in the historical order information as preset goods; and determining the historical inquiry characteristics corresponding to the historical recommended commodity and the historical inquiry characteristics corresponding to the historical order information as inquiry characteristic labels of preset commodities; the training module (not shown in the figure) is used for training the preset model according to the preset commodity and the inquiry feature label to obtain the recommended model.

In an embodiment, the screening module 302 is specifically configured to determine, as the second commodity, a preset commodity whose heat information matches the preset heat information; the heat information comprises recommended times of preset commodities, searching times of the preset commodities, sales amount of the preset commodities and/or evaluation amount of the preset commodities.

In an embodiment, the medical commodity recommending device further includes: a time determining module (not shown in the figure), which is used for determining a preset period to which the current time belongs; and calculating a target time period corresponding to the history time of the same ratio or the ring ratio according to the time minimum value and the time maximum value of the preset time period.

In an embodiment, the filtering module 302 is further configured to delete the recommended merchandise if the user characteristic corresponding to the inquiry log information does not conform to the applicable condition information of the recommended merchandise.

In an embodiment, the medical commodity recommending device further includes: a location determining module (not shown in the figure) for acquiring location information of the target user; the screening module 302 is further configured to determine a target merchant located in a distribution location range corresponding to the positioning information; if the stock quantity of the recommended commodity corresponding to the target merchant does not accord with the demand quantity of the recommended commodity, deleting the recommended commodity; the communication module 303 is further configured to send navigation information of the target merchant to a user side of the target user in response to an order confirmation instruction of the recommended commodity if the inventory amount of the recommended commodity corresponding to the target merchant meets the demand amount of the recommended commodity.

In an embodiment, the medical commodity recommending device further includes: the ordering module (not shown in the figure) is used for ordering the recommended goods according to a plurality of preset ordering strategies to obtain a plurality of recommendation lists of the recommended goods; the communication module 303 is further configured to send the plurality of recommendation lists to a doctor end, so that the doctor end responds to a display instruction of the target recommendation list and displays commodity information of the recommended commodities according to the order of the target recommendation list.

In an embodiment, the medical commodity recommending device further includes: the scoring module (not shown in the figure) is used for determining the weight score of the first commodity in the recommended commodity according to the matching degree of the first commodity and the inquiry feature and the first weight under the condition that the preset ranking strategy is arranged according to the weight score; determining a weight score of a second commodity in the recommended commodity according to the heat information and the second weight; and determining the weight score of the third commodity in the recommended commodity according to the priority of the target activity and the third weight.

The specific limitation of the recommendation device for medical goods may be referred to the limitation of the recommendation method for medical goods hereinabove, and will not be described herein. The respective modules in the medical commodity recommending apparatus described above may be implemented in whole or in part by software, hardware, or a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the steps of when executing the computer program: responding to a commodity display request of a target user generated by triggering of a doctor end, and analyzing inquiry log information of the target user corresponding to the commodity display request; screening preset commodities according to the inquiry log information to determine a first commodity; screening the preset commodity according to the heat information of the preset commodity in the target period of the current moment to determine a second commodity; if the inquiry log information accords with the condition information of the target activity, determining an activity commodity corresponding to the target activity as a third commodity; performing duplicate removal processing on the first commodity, the second commodity and the third commodity to determine recommended commodity; and sending the commodity information of the recommended commodity to a doctor end so that the doctor end displays the commodity information of the recommended commodity.

In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of: responding to a commodity display request of a target user generated by triggering of a doctor end, and analyzing inquiry log information of the target user corresponding to the commodity display request; screening preset commodities according to the inquiry log information to determine a first commodity; screening the preset commodity according to the heat information of the preset commodity in the target period of the current moment to determine a second commodity; if the inquiry log information accords with the condition information of the target activity, determining an activity commodity corresponding to the target activity as a third commodity; performing duplicate removal processing on the first commodity, the second commodity and the third commodity to determine recommended commodity; and sending the commodity information of the recommended commodity to a doctor end so that the doctor end displays the commodity information of the recommended commodity.

In one embodiment, a computer device is provided, which may be a client, the internal structure of which is shown in FIG. 4. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is for communicating with an external server via a network connection. The computer program is executed by a processor to perform the functions or steps of a method of recommending medical goods.

It should be noted that, the functions or steps that can be implemented by the computer readable storage medium or the computer device may correspond to the descriptions related to the recommended method of the medical commodity in the foregoing method embodiment, and are not described herein one by one for avoiding repetition.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.

It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions.

The foregoing embodiments are merely for illustrating the technical aspects of the present invention, and not for limiting the same, and although the present invention has been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that the technical aspects described in the foregoing embodiments may be modified or some of the technical features thereof may be replaced by others; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention.

Claims

1. A method of recommending medical goods, comprising:

responding to a commodity display request of a target user generated by triggering of a doctor side, and analyzing inquiry log information of the target user corresponding to the commodity display request;

and sending the commodity information of the recommended commodity to the doctor end so that the doctor end displays the commodity information of the recommended commodity.

2. The method for recommending medical goods according to claim 1, wherein the screening preset goods according to the inquiry log information to determine the first goods comprises:

inputting the inquiry features into a recommendation model to obtain the first commodity and the matching degree of the first commodity and the inquiry features;

3. The method of recommending medical goods according to claim 2, wherein prior to said entering said inquiry feature into a recommendation model, said method further comprises:

acquiring historical recommended commodities aiming at different target users and historical order information of the different target users sent by the doctor side;

determining the historical recommended commodity and the order commodity in the historical order information as the preset commodity;

determining the historical inquiry characteristics corresponding to the historical recommended commodity and the historical inquiry characteristics corresponding to the historical order information as inquiry characteristic labels of the preset commodity;

training a preset model according to the preset commodity and the inquiry feature tag to obtain the recommended model.

4. The medical commodity recommendation method according to claim 1, wherein the screening the preset commodity according to the heat information of the preset commodity in the target period to which the current time belongs includes:

determining the preset commodity with the heat information conforming to the preset heat information as the second commodity;

the heat information comprises recommended times of the preset commodity, searching times of the preset commodity, sales amount of the preset commodity and/or evaluation amount of the preset commodity.

5. The method of recommending medical goods according to any one of claims 1 to 4, further comprising:

acquiring positioning information of the target user;

if the stock quantity of the recommended commodity corresponding to the target merchant does not meet the demand quantity of the recommended commodity, deleting the recommended commodity;

and if the stock quantity of the recommended commodity corresponding to the target merchant accords with the demand quantity of the recommended commodity, responding to an order confirmation instruction of the recommended commodity, and sending the navigation information of the target merchant to a user side of the target user.

6. The method of recommending medical goods according to any one of claims 1 to 4, further comprising:

and sending the plurality of recommendation lists to the doctor terminal so that the doctor terminal responds to the display instruction of the target recommendation list and displays commodity information of the recommended commodities according to the sequence of the target recommendation list.

7. The method of claim 6, wherein prior to ranking the recommended items according to a plurality of preset ranking strategies, the method further comprises:

under the condition that the preset ordering strategy is arranged according to the weight scores, determining the weight scores of the first commodity in the recommended commodity according to the matching degree of the first commodity and the inquiry feature and the first weight;

determining a weight score of the second commodity in the recommended commodity according to the heat information and the second weight;

and determining a weight score of the third commodity in the recommended commodity according to the priority of the target activity and the third weight.

8. A medical commodity recommending apparatus, comprising:

the analysis module is used for responding to a commodity display request of a target user generated by the triggering of a doctor end and analyzing inquiry log information of the target user corresponding to the commodity display request;

the screening module is used for screening preset commodities according to the inquiry log information to determine a first commodity; the method comprises the steps of,

screening the preset commodity according to the heat information of the preset commodity in the target period of the current moment to determine a second commodity; the method comprises the steps of,

If the inquiry log information accords with the condition information of the target activity, determining an activity commodity corresponding to the target activity as a third commodity; the method comprises the steps of,

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the recommendation method of a medical article according to any one of claims 1 to 7 when the computer program is executed by the processor.

10. A computer-readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the steps of the recommendation method of medical articles according to any one of claims 1 to 7.