CN115757982A - Medical instrument recommendation method and system based on artificial intelligence - Google Patents
Medical instrument recommendation method and system based on artificial intelligence Download PDFInfo
- Publication number
- CN115757982A CN115757982A CN202111031007.6A CN202111031007A CN115757982A CN 115757982 A CN115757982 A CN 115757982A CN 202111031007 A CN202111031007 A CN 202111031007A CN 115757982 A CN115757982 A CN 115757982A
- Authority
- CN
- China
- Prior art keywords
- target user
- user
- behavior data
- preference
- dimensional matrix
- 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.)
- Pending
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application discloses a medical instrument recommendation method and system based on artificial intelligence, and the method comprises the following steps: the behavior data of a target user are collected, the behavior data are preprocessed on the basis of artificial intelligence, a two-dimensional matrix preferred by the target user is obtained according to behavior analysis methods of different applications on the preprocessed behavior data, one dimension of the two-dimensional matrix is a target user list, the other dimension of the two-dimensional matrix is a medical apparatus list, the value of the two-dimensional matrix is the preference of the target user on the medical apparatus, adjacent neighbor users are determined on the basis of the preference of the target user on the medical apparatus, and the preference of the adjacent neighbor users is recommended to the target user. According to the method and the device, the effectiveness of recommendation is improved, the accuracy of recommending medical instruments is improved and the recommended medical instruments are more personalized based on the intelligent recommendation technology of artificial intelligence.
Description
Technical Field
The application relates to the technical field of data processing, in particular to a medical instrument recommendation method based on artificial intelligence.
Background
With the rapid development and widespread application of electronic commerce, medical device recommendation has become an important research field. At present, content recommendation mainly adopts a collaborative filtering recommendation mode, and the technical scheme of the collaborative filtering recommendation mode mainly comprises scoring marking, proximity selection and recommendation generation. Wherein, the score indication, i.e. the input data of the traditional collaborative filtering recommendation algorithm, is an m × n user-item score matrix. And (3) the adjacent selection, namely the recommendation principle of the collaborative filtering algorithm is to search for a neighboring user similar to the target user and generate recommendation for the target user through the evaluation of the neighboring user. The selection method of the neighbor users is as follows: and calculating the similarity between the target user and all other users in the recommendation system, and sequentially selecting the previous K most similar users as a neighbor set of the target user from large to small according to the similarity ranking. The recommendation is generated, namely, a basic assumption of the collaborative filtering algorithm is that users with similar preferences give similar scores to the same item, so that after a neighbor set of a target user is generated, the score of the target user for an unscored item can be predicted according to the scores of the users in the neighbor set. Therefore, the collaborative filtering technology has been widely applied and has great success in the recommendation system, but with the development and popularization of the internet, the explosive increase of the number of users, the number of commodities and network resources, the increase of the complexity of the site structure, and the continuous upgrade of the network information security, the collaborative filtering recommendation system also faces the following problems and challenges: data sparseness, cold start problems, scalability problems, robustness problems, implicit preference discovery, and the like.
Disclosure of Invention
In view of this, the present application provides a medical device recommendation method based on artificial intelligence, which can improve the effectiveness of recommendation, the accuracy of recommending medical devices, and make the recommended medical devices more personalized based on an intelligent recommendation technology of artificial intelligence.
The application provides a medical instrument recommendation method based on artificial intelligence, which comprises the following steps: collecting behavior data of a target user; preprocessing the behavior data based on artificial intelligence; obtaining a two-dimensional matrix preferred by the target user according to behavior analysis methods applied differently to the preprocessed behavior data, wherein one dimension of the two-dimensional matrix is a target user list, the other dimension of the two-dimensional matrix is a medical apparatus list, and the value of the two-dimensional matrix is the preference of the target user to the medical apparatus; and determining adjacent neighbor users based on the preference of the target user for the medical equipment, and recommending the preference of the adjacent neighbor users to the target user.
Preferably, the preprocessing the behavior data based on artificial intelligence includes: and carrying out noise reduction and normalization processing on the behavior data.
Preferably, the noise reduction processing of the behavior data includes: and filtering out noise in the behavior data through a data mining algorithm.
Preferably, normalizing the behavior data includes: and unifying the behavior data of each behavior in the same value range through normalization processing.
Preferably, the determining of the adjacent neighbor users based on the target user's preference for the medical instrument comprises: and calculating the similarity between the users by taking the preference of the users to the medical equipment as a vector, and determining the users with the similarity meeting the preset conditions with the target user as adjacent neighbor users.
An artificial intelligence based medical device recommendation system comprising: the collection module is used for collecting behavior data of a target user; the processing module is used for preprocessing the behavior data based on artificial intelligence; the analysis module is used for obtaining a two-dimensional matrix preferred by the target user according to behavior analysis methods applied differently to the preprocessed behavior data, wherein one dimension of the two-dimensional matrix is a target user list, the other dimension of the two-dimensional matrix is a medical instrument list, and the value of the two-dimensional matrix is the preference of the target user on the medical instrument;
and the recommending module is used for determining adjacent neighbor users based on the preference of the target user for the medical equipment and recommending the preference of the adjacent neighbor users to the target user.
Preferably, the processing module is specifically configured to: and carrying out noise reduction and normalization processing on the behavior data.
Preferably, the processing module comprises: and the denoising unit is used for filtering noise in the behavior data through a data mining algorithm.
Preferably, the processing module further comprises: and the normalization unit is used for unifying the behavior data of each behavior in the same value range through normalization processing.
Preferably, the recommendation module is specifically configured to:
the method comprises the steps of calculating similarity between users by taking the preference of the users to medical equipment as a vector, determining the users with the similarity meeting preset conditions with a target user as adjacent neighbor users, and recommending the preference of the adjacent neighbor users to the target user.
In summary, the application discloses a medical device recommendation method based on artificial intelligence, when information needs to be recommended to a user, behavior data of a target user is collected first, then the behavior data is preprocessed based on artificial intelligence, the preprocessed behavior data is subjected to behavior analysis methods according to different applications, a two-dimensional matrix of target user preferences is obtained, wherein one dimension of the two-dimensional matrix is a target user list, the other dimension of the two-dimensional matrix is a medical device list, the value of the two-dimensional matrix is the target user preference for a medical device, and finally an adjacent neighbor user is determined based on the target user preference for the medical device, and the preferences of the adjacent neighbor user is recommended to the target user. According to the method and the device, the effectiveness of recommendation is improved, the accuracy of recommending medical equipment is improved and the recommended medical equipment is more personalized based on the intelligent recommendation technology of artificial intelligence.
Detailed Description
The following will clearly and completely describe the technical solutions in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
The application discloses a medical instrument recommendation method based on artificial intelligence, which comprises the following steps:
collecting behavior data of a target user; and (3) recommending contents based on artificial intelligence, and adopting a collaborative filtering recommendation algorithm and a deep machine learning technology. When recommending, user preferences need to be collected, rules are found from the behaviors and preferences of the users, and recommendations are given based on the rules, and how to collect preference information of the users becomes the most basic determining factor of the recommendation effect of the system.
The user has many ways to provide own preference information to the system, and different applications may be quite different, different behaviors are grouped, generally divided into 'viewing' and 'purchasing' and the like, and then different user/medical apparatus similarities are calculated based on different behaviors; and then weighting the different behaviors according to the degree of reflecting the user preference to obtain the overall preference of the user for the medical instrument. Generally, the explicit user feedback is more than the implicit weight, but is sparse, and after all, the number of users who perform the display feedback is small; while the user's preferences are reflected to a greater extent than the "view", "buy" behavior, but this also varies from application to application.
Therefore, when content recommendation is performed, behavior data of a target user of the mobile phone is needed first, wherein the target user refers to a user who needs to be subjected to content recommendation.
Preprocessing the behavior data based on artificial intelligence;
after the mobile phone reaches the behavior data of the target user, the behavior data is further preprocessed by artificial intelligence.
It should be noted that the artificial intelligence disciplines are cross disciplines and edge disciplines relating to disciplines such as mathematics, computer disciplines, cybernetics, psychology and philosophy, and the application fields thereof include problem solving, expert systems, machine learning, pattern recognition, automatic theorem proving, natural language understanding, artificial neural networks, intelligent retrieval and the like. The artificial intelligence is mainly used for information retrieval by intelligent retrieval methods based on ontology, neural network, genetic algorithm, natural language understanding, ID3 algorithm and the like.
Obtaining a two-dimensional matrix preferred by a target user according to behavior analysis methods applied differently to the preprocessed behavior data, wherein one dimension of the two-dimensional matrix is a target user list, the other dimension of the two-dimensional matrix is a medical apparatus list, and the value of the two-dimensional matrix is the preference of the target user to the medical apparatus; after preprocessing the behavior data, according to different applied behavior analysis methods, grouping or weighting processing can be selected, and then we can obtain a two-dimensional matrix of user preferences, one dimension is a user list, the other dimension is a medical instrument list, and the value is the user preference for the medical instrument, and is generally a floating point value of [0,1] or [ -1,1 ].
And determining adjacent neighbor users based on the preference of the target user for the medical equipment, and recommending the preference of the adjacent neighbor users to the target user. After the user behavior has been analyzed to obtain the user preferences, similar users and medical devices may be calculated according to the user preferences, and then the target user may be recommended based on the similar users or the medical devices.
In summary, in the above embodiments, when information needs to be recommended to a user, behavior data of a target user is collected first, then the behavior data is preprocessed based on artificial intelligence, a two-dimensional matrix of preferences of the target user is obtained for the preprocessed behavior data according to behavior analysis methods applied differently, where one dimension of the two-dimensional matrix is a target user list, the other dimension of the two-dimensional matrix is a medical apparatus list, and a value of the two-dimensional matrix is a preference of the target user for the medical apparatus, and finally an adjacent neighbor user is determined based on the preference of the target user for the medical apparatus, and the preference of the adjacent neighbor user is recommended to the target user. According to the method and the device, the effectiveness of recommendation is improved, the accuracy of recommending medical instruments is improved and the recommended medical instruments are more personalized based on the intelligent recommendation technology of artificial intelligence.
Example 2
The application discloses a medical instrument recommendation method based on artificial intelligence, which comprises the following steps:
collecting behavior data of a target user; and (3) recommending contents based on artificial intelligence, and adopting a collaborative filtering recommendation algorithm and a deep machine learning technology. When recommending, user preferences need to be collected, rules are found from the behaviors and preferences of the users, and recommendations are given based on the rules, and how to collect preference information of the users becomes the most basic determining factor of the recommendation effect of the system.
The user has many ways to provide own preference information for the system, and different applications may be very different, different behaviors are grouped, generally divided into 'viewing' and 'purchasing' and the like, and then different user/medical apparatus similarities are calculated based on different behaviors; and then weighting the different behaviors according to the degree of reflecting the user preference to obtain the overall preference of the user on the medical instrument. Generally, the explicit user feedback is more than the implicit weight, but is sparse, and after all, the number of users who perform the display feedback is small; while the user's preferences are reflected to a greater extent than the "view", "buy" behavior, but this also varies from application to application.
Therefore, when content recommendation is performed, behavior data of a target user of the mobile phone is needed first, wherein the target user refers to a user who needs to be subjected to content recommendation.
Performing noise reduction and normalization processing on the behavior data based on artificial intelligence; after the mobile phone reaches the behavior data of the target user, the behavior data is further preprocessed by artificial intelligence.
The most core work of preprocessing the behavior data is noise reduction and normalization. It should be noted that the artificial intelligence disciplines are cross disciplines and edge disciplines relating to disciplines such as mathematics, computer disciplines, cybernetics, psychology and philosophy, and the application fields thereof include problem solving, expert systems, machine learning, pattern recognition, automatic theorem proving, natural language understanding, artificial neural networks, intelligent retrieval and the like. The artificial intelligence is mainly used for information retrieval by intelligent retrieval methods based on ontology, neural network, genetic algorithm, natural language understanding, ID3 algorithm and the like.
Obtaining a two-dimensional matrix preferred by a target user according to behavior analysis methods applied differently to the preprocessed behavior data, wherein one dimension of the two-dimensional matrix is a target user list, the other dimension of the two-dimensional matrix is a medical apparatus list, and the value of the two-dimensional matrix is the preference of the target user to the medical apparatus; after preprocessing the behavior data, according to different applied behavior analysis methods, grouping or weighting processing can be selected, and then we can obtain a two-dimensional matrix of user preferences, one dimension is a user list, the other dimension is a medical instrument list, and the value is the user preference for the medical instrument, and is generally a floating point value of [0,1] or [ -1,1 ].
And determining adjacent neighbor users based on the preference of the target user for the medical equipment, and recommending the preference of the adjacent neighbor users to the target user. After the user behavior has been analyzed to obtain the user preferences, similar users and medical devices may be calculated according to the user preferences, and then the target user may be recommended based on the similar users or the medical devices.
In summary, in the above embodiments, when information needs to be recommended to a user, behavior data of a target user is collected first, noise reduction and normalization processing is performed on the behavior data based on artificial intelligence, a two-dimensional matrix of preferences of the target user is obtained for the behavior data after preprocessing according to behavior analysis methods applied differently, where one dimension of the two-dimensional matrix is a target user list, the other dimension is a medical apparatus list, and a value is a preference of the target user for a medical apparatus, and finally an adjacent neighbor user is determined based on the preference of the target user for the medical apparatus, and the preference of the adjacent neighbor user is recommended to the target user. According to the method and the device, the effectiveness of recommendation is improved, the accuracy of recommending medical equipment is improved and the recommended medical equipment is more personalized based on the intelligent recommendation technology of artificial intelligence.
Example 3
The medical instrument recommendation method based on artificial intelligence disclosed by the application can actually comprise the following steps of:
collecting behavior data of a target user; and (3) recommending contents based on artificial intelligence, and adopting a collaborative filtering recommendation algorithm and a deep machine learning technology. When recommending, user preferences need to be collected, rules are found from the behaviors and preferences of the users, and recommendations are given based on the rules, and how to collect preference information of the users becomes the most basic determining factor of the recommendation effect of the system.
The user has many ways to provide own preference information to the system, and different applications may be quite different, different behaviors are grouped, generally divided into 'viewing' and 'purchasing' and the like, and then different user/medical apparatus similarities are calculated based on different behaviors; and then weighting the different behaviors according to the degree of reflecting the user preference to obtain the overall preference of the user for the medical instrument. Generally, the explicit user feedback is more than the implicit weight, but is sparse, and after all, the number of users who perform the display feedback is small; while the user's preferences are reflected to a greater extent with respect to "view", "buy" behavior, but this also varies from application to application.
Therefore, when content recommendation is performed, behavior data of a target user of the mobile phone is needed first, wherein the target user refers to a user who needs to be performed content recommendation.
Filtering noise in the behavior data through a data mining algorithm; after the mobile phone reaches the behavior data of the target user, the behavior data is further preprocessed by artificial intelligence. The most central work for preprocessing the behavior data is noise reduction and normalization. The user behavior data is generated by the user in the application using process, a large amount of noise and misoperation of the user can exist, so that the noise in the behavior data can be filtered out through a classical data mining algorithm, and the analysis can be more accurate.
Behavior data of each behavior are unified in the same value range through normalization processing; how to unify the data of each behavior in the same value range so that the overall preference obtained by weighting and summing is more accurate needs to normalize the behavior data.
Obtaining a two-dimensional matrix preferred by a target user according to behavior analysis methods applied differently to the preprocessed behavior data, wherein one dimension of the two-dimensional matrix is a target user list, the other dimension of the two-dimensional matrix is a medical apparatus list, and the value of the two-dimensional matrix is the preference of the target user to the medical apparatus; after preprocessing the behavior data, according to different applied behavior analysis methods, grouping or weighting processing can be selected, and then we can obtain a two-dimensional matrix of user preferences, one dimension is a user list, the other dimension is a medical instrument list, and the value is the user preference for the medical instrument, and is generally a floating point value of [0,1] or [ -1,1 ].
And calculating the similarity between the users by taking the preference of the users to the medical equipment as a vector, and determining the users with the similarity meeting the preset conditions with the target user as adjacent neighbor users.
When the user behavior has been analyzed to obtain user preferences, similar users and medical devices may be calculated according to the user preferences, and then recommendations are made based on the similar users or medical devices, which are two branches of the most typical CF: user-based CFs and medical appliance-based CFs. And (3) calculating the similarity: several basic methods are vector-based, that is, the distance between two vectors is calculated, and the closer the distance is, the greater the similarity is. In the recommended scenario, in the two-dimensional matrix of user-modality preferences, we can calculate the similarity between users by using the preference of one user for all modalities as a vector, or calculate the similarity between modalities by using the preference of all users for a certain modality as a vector. Calculation of similar neighbors: a fixed number of neighbors and a similarity threshold based neighbor. The computed recommendations are then used to find neighboring neighbor users based on the user's preferences for the medical device, and then recommendations that the neighbor users like are then provided to the current user. And calculating to obtain a sorted medical instrument list as recommendation by using the preference of a user to all medical instruments as a vector to calculate the similarity between the users, predicting the medical instruments which are not concerned by the current user and have no preference according to the similarity weight of the neighbors and the preference of the neighbors to the medical instruments after finding the B neighbors.
In summary, in the embodiments, the content is recommended by the collaborative filtering method based on the artificial intelligence algorithm, the first-layer learning network is constructed by establishing the machine learning model, the second-layer learning network is constructed by taking the learning label of the first-layer network as an input, and by analogy, a deeper learning network is constructed. The automatic crawler acquires big data, and the big data are trained through the model, so that the corrected learning result is continuously improved, and the learning efficiency is improved. Analyzing and researching the target user behavior data, comparing the target user behavior data with neighbors marked with the same preferences, and generating recommendation to the target user according to the preferences of the neighbors of the target user. The effectiveness of recommendation is improved, the accuracy of recommending medical instruments is improved, and the recommended medical instruments are more personalized.
Example 4
The application discloses a medical instrument recommendation system based on artificial intelligence can include:
the collecting module is used for collecting behavior data of a target user;
and (3) recommending contents based on artificial intelligence, and adopting a collaborative filtering recommendation algorithm and a deep machine learning technology. When recommending, user preferences need to be collected, rules need to be found from behaviors and preferences of users, and recommendations are given based on the rules, and how to collect preference information of the users becomes the most basic determining factor of system recommendation effect.
The user has many ways to provide own preference information to the system, and different applications may be quite different, different behaviors are grouped, generally divided into 'viewing' and 'purchasing' and the like, and then different user/medical apparatus similarities are calculated based on different behaviors; and then, weighting operation is carried out, and the different behaviors are weighted according to the degree of reflecting the user preference, so that the overall preference of the user for the medical instrument is obtained. Generally, the explicit user feedback is more than the implicit weight, but is sparse, and after all, the number of users who perform the display feedback is small; while the user's preferences are reflected to a greater extent with respect to "view", "buy" behavior, but this also varies from application to application.
Therefore, when content recommendation is performed, behavior data of a target user of the mobile phone is needed first, wherein the target user refers to a user who needs to be performed content recommendation.
The processing module is used for preprocessing the behavior data based on artificial intelligence; after the mobile phone reaches the behavior data of the target user, the behavior data is further preprocessed by artificial intelligence. It should be noted that the artificial intelligence disciplines are cross disciplines and edge disciplines related to the disciplines of mathematics, computer science, cybernetics, psychology, philosophy, etc., and the application fields thereof include problem solving, expert system, machine learning, pattern recognition, automatic theorem proving, natural language understanding, artificial neural network, intelligent retrieval, etc. The artificial intelligence is mainly used for information retrieval by intelligent retrieval methods based on ontology, neural network, genetic algorithm, natural language understanding, ID3 algorithm and the like.
The analysis module is used for obtaining a two-dimensional matrix preferred by the target user according to behavior analysis methods of different applications for the preprocessed behavior data, wherein one dimension of the two-dimensional matrix is a target user list, the other dimension of the two-dimensional matrix is a medical apparatus list, and the value of the two-dimensional matrix is the preference of the target user for the medical apparatus; after preprocessing the behavior data, grouping or weighting can be selected according to different applied behavior analysis methods, and then we can obtain a two-dimensional matrix of user preferences, where one dimension is a user list, the other dimension is a medical device list, and the value is the user preference for the medical device, and is generally a floating point value of [0,1] or [ -1,1 ].
And the recommending module is used for determining the adjacent neighbor users based on the preference of the target user for the medical apparatus and recommending the preference of the adjacent neighbor users to the target user.
After the user behavior has been analyzed to obtain the user preferences, similar users and medical devices may be calculated according to the user preferences, and then the target user may be recommended based on the similar users or the medical devices.
In summary, in the above embodiments, when information needs to be recommended to a user, behavior data of a target user is collected first, then the behavior data is preprocessed based on artificial intelligence, a two-dimensional matrix of preferences of the target user is obtained for the preprocessed behavior data according to behavior analysis methods applied differently, where one dimension of the two-dimensional matrix is a target user list, the other dimension of the two-dimensional matrix is a medical apparatus list, and a value of the two-dimensional matrix is a preference of the target user for the medical apparatus, and finally an adjacent neighbor user is determined based on the preference of the target user for the medical apparatus, and the preference of the adjacent neighbor user is recommended to the target user. According to the method and the device, the effectiveness of recommendation is improved, the accuracy of recommending medical equipment is improved and the recommended medical equipment is more personalized based on the intelligent recommendation technology of artificial intelligence.
Example 5
The application discloses a medical instrument recommendation system based on artificial intelligence can include:
the collection module is used for collecting behavior data of a target user; and (3) recommending contents based on artificial intelligence, and adopting a collaborative filtering recommendation algorithm and a deep machine learning technology. When recommending, user preferences need to be collected, rules are found from the behaviors and preferences of the users, and recommendations are given based on the rules, and how to collect preference information of the users becomes the most basic determining factor of the recommendation effect of the system.
The user has many ways to provide own preference information to the system, and different applications may be quite different, different behaviors are grouped, generally divided into 'viewing' and 'purchasing' and the like, and then different user/medical apparatus similarities are calculated based on different behaviors; and then weighting the different behaviors according to the degree of reflecting the user preference to obtain the overall preference of the user for the medical instrument. Generally, the weight of explicit user feedback is larger than that of implicit user feedback, but the explicit user feedback is sparse, and after all, the number of users who perform display feedback is small; while the user's preferences are reflected to a greater extent than the "view", "buy" behavior, but this also varies from application to application.
Therefore, when content recommendation is performed, behavior data of a target user of the mobile phone is needed first, wherein the target user refers to a user who needs to be performed content recommendation.
The processing module is used for carrying out noise reduction and normalization processing on the behavior data based on artificial intelligence; after the mobile phone reaches the behavior data of the target user, the behavior data is further preprocessed by artificial intelligence.
The most central work for preprocessing the behavior data is noise reduction and normalization. It should be noted that the artificial intelligence disciplines are cross disciplines and edge disciplines relating to disciplines such as mathematics, computer disciplines, cybernetics, psychology and philosophy, and the application fields thereof include problem solving, expert systems, machine learning, pattern recognition, automatic theorem proving, natural language understanding, artificial neural networks, intelligent retrieval and the like. The artificial intelligence is mainly used for information retrieval by intelligent retrieval methods based on ontology, neural network, genetic algorithm, natural language understanding, ID3 algorithm and the like.
The analysis module is used for obtaining a two-dimensional matrix preferred by the target user according to behavior analysis methods of different applications for the preprocessed behavior data, wherein one dimension of the two-dimensional matrix is a target user list, the other dimension of the two-dimensional matrix is a medical apparatus list, and the value of the two-dimensional matrix is the preference of the target user for the medical apparatus; after preprocessing the behavior data, according to different applied behavior analysis methods, grouping or weighting processing can be selected, and then we can obtain a two-dimensional matrix of user preferences, one dimension is a user list, the other dimension is a medical instrument list, and the value is the user preference for the medical instrument, and is generally a floating point value of [0,1] or [ -1,1 ].
And the recommending module is used for determining the adjacent neighbor users based on the preference of the target user for the medical apparatus and recommending the preference of the adjacent neighbor users to the target user.
After the user behavior has been analyzed to obtain the user preferences, similar users and medical devices may be calculated according to the user preferences, and then the target user may be recommended based on the similar users or the medical devices.
In summary, in the above embodiments, when information needs to be recommended to a user, behavior data of a target user is collected first, noise reduction and normalization processing is performed on the behavior data based on artificial intelligence, a two-dimensional matrix of preferences of the target user is obtained for the behavior data after preprocessing according to behavior analysis methods applied differently, where one dimension of the two-dimensional matrix is a target user list, the other dimension is a medical apparatus list, and a value is a preference of the target user for a medical apparatus, and finally an adjacent neighbor user is determined based on the preference of the target user for the medical apparatus, and the preference of the adjacent neighbor user is recommended to the target user. According to the method and the device, the effectiveness of recommendation is improved, the accuracy of recommending medical equipment is improved and the recommended medical equipment is more personalized based on the intelligent recommendation technology of artificial intelligence.
Example 6
Medical instrument recommendation based on artificial intelligence
The system may include: the collection module is used for collecting behavior data of a target user; and (3) recommending contents based on artificial intelligence, and adopting a collaborative filtering recommendation algorithm and a deep machine learning technology. When recommending, user preferences need to be collected, rules need to be found from behaviors and preferences of users, and recommendations are given based on the rules, and how to collect preference information of the users becomes the most basic determining factor of system recommendation effect.
The user has many ways to provide own preference information to the system, and different applications may be quite different, different behaviors are grouped, generally divided into 'viewing' and 'purchasing' and the like, and then different user/medical apparatus similarities are calculated based on different behaviors; and then weighting the different behaviors according to the degree of reflecting the user preference to obtain the overall preference of the user for the medical instrument. Generally, the weight of explicit user feedback is larger than that of implicit user feedback, but the explicit user feedback is sparse, and after all, the number of users who perform display feedback is small; while the user's preferences are reflected to a greater extent with respect to "view", "buy" behavior, but this also varies from application to application.
Therefore, when content recommendation is performed, behavior data of a target user of the mobile phone is needed first, wherein the target user refers to a user who needs to be subjected to content recommendation.
The denoising unit is used for filtering noise in the behavior data through a data mining algorithm; after the mobile phone reaches the behavior data of the target user, the behavior data is further preprocessed by artificial intelligence.
The most central work for preprocessing the behavior data is noise reduction and normalization. The user behavior data is generated by the user in the application using process, a large amount of noise and misoperation of the user can exist, and therefore the noise in the behavior data can be filtered through a classical data mining algorithm, and the analysis can be more accurate.
The normalization unit is used for unifying the behavior data of each behavior in the same value range through normalization processing;
how to unify the data of each behavior in the same value range so that the overall preference obtained by weighting and summing is more accurate needs to normalize the behavior data.
The analysis module is used for obtaining a two-dimensional matrix preferred by the target user according to behavior analysis methods of different applications for the preprocessed behavior data, wherein one dimension of the two-dimensional matrix is a target user list, the other dimension of the two-dimensional matrix is a medical apparatus list, and the value of the two-dimensional matrix is the preference of the target user for the medical apparatus; after preprocessing the behavior data, according to different applied behavior analysis methods, grouping or weighting processing can be selected, and then we can obtain a two-dimensional matrix of user preferences, one dimension is a user list, the other dimension is a medical instrument list, and the value is the user preference for the medical instrument, and is generally a floating point value of [0,1] or [ -1,1 ].
And the recommendation module is used for calculating the similarity between the users by taking the preference of the users to the medical equipment as a vector, and determining the user with the similarity meeting the preset condition with the target user as an adjacent neighbor user. When the user behavior has been analyzed to obtain user preferences, similar users and medical devices may be calculated according to the user preferences, and then recommendations are made based on the similar users or medical devices, which are two branches of the most typical CF: user-based CFs and medical appliance-based CFs. And (3) calculating the similarity: several basic methods are vector-based, which means that the distance between two vectors is calculated, and the closer the distance is, the greater the similarity is. In the recommended scenario, in the two-dimensional matrix of user-modality preferences, we can calculate the similarity between users by using the preferences of one user for all modalities as a vector, or calculate the similarity between modalities by using the preferences of all users for a certain modality as a vector. Calculation of similar neighbors: a fixed number of neighbors and a similarity threshold based neighbor. The computed recommendations are then used to find neighboring neighbor users based on the user's preferences for the medical device, and then recommendations that the neighbor users like are then provided to the current user. And calculating to obtain a sorted medical instrument list as recommendation by using the preference of a user to all medical instruments as a vector to calculate the similarity between the users, predicting the medical instruments which are not concerned by the current user and have no preference according to the similarity weight of the neighbors and the preference of the neighbors to the medical instruments after finding the B neighbors.
In summary, in the embodiments, the content is recommended by the collaborative filtering method based on the artificial intelligence algorithm, the first-layer learning network is constructed by establishing the machine learning model, the second-layer learning network is constructed by taking the learning label of the first-layer network as an input, and by analogy, a deeper learning network is constructed. The automatic crawler acquires big data, and the big data are trained through the model, so that the learning result is continuously improved and corrected, and the learning efficiency is improved. Analyzing and researching the target user behavior data, comparing the target user behavior data with neighbors marked with the same preferences, and generating recommendation to the target user according to the preferences of the neighbors of the target user. The recommendation effectiveness and the recommendation accuracy of medical instruments are improved, and the recommended medical instruments have individuality.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed in the embodiment corresponds to the method disclosed in the embodiment, so that the description is simple, and the relevant points can be referred to the description of the method part. Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the technical solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A medical device recommendation method based on artificial intelligence is characterized by comprising the following steps: collecting behavior data of a target user; preprocessing the behavior data based on artificial intelligence; obtaining a two-dimensional matrix preferred by the target user according to behavior analysis methods applied differently to the preprocessed behavior data, wherein one dimension of the two-dimensional matrix is a target user list, the other dimension of the two-dimensional matrix is a medical apparatus list, and the value of the two-dimensional matrix is the preference of the target user to the medical apparatus; and determining adjacent neighbor users based on the preference of the target user for the medical equipment, and recommending the preference of the adjacent neighbor users to the target user.
2. The method of claim 1, wherein the artificial intelligence based preprocessing of the behavioral data comprises: and performing noise reduction and normalization processing on the behavior data.
3. The method of claim 2, wherein noise-reducing the behavioral data comprises: and filtering out noise in the behavior data through a data mining algorithm.
4. The method of claim 2, wherein normalizing the behavior data comprises: and unifying the behavior data of each behavior in the same value range through normalization processing.
5. The method of claim 1, wherein determining neighboring neighbor users based on target user preferences for medical instruments comprises: and calculating the similarity between the users by taking the preference of the users to the medical equipment as a vector, and determining the users with the similarity meeting preset conditions with the target user as adjacent neighbor users.
6. An artificial intelligence based medical device recommendation system, comprising: the collection module is used for collecting behavior data of a target user; the processing module is used for preprocessing the behavior data based on artificial intelligence; the analysis module is used for obtaining a two-dimensional matrix preferred by the target user according to behavior analysis methods applied differently to the preprocessed behavior data, wherein one dimension of the two-dimensional matrix is a target user list, the other dimension of the two-dimensional matrix is a medical instrument list, and the value of the two-dimensional matrix is the preference of the target user on the medical instrument; and the recommending module is used for determining adjacent neighbor users based on the preference of the target user for the medical equipment and recommending the preference of the adjacent neighbor users to the target user.
7. The system of claim 6, wherein the processing module is specifically configured to: and performing noise reduction and normalization processing on the behavior data.
8. The system of claim 7, wherein the processing module comprises: and the denoising unit is used for filtering noise in the behavior data through a data mining algorithm.
9. The system of claim 7, wherein the processing module further comprises: and the normalization unit is used for unifying the behavior data of each behavior in the same value range through normalization processing.
10. The system of claim 6, wherein the recommendation module is specifically configured to: the method comprises the steps of taking the preference of a user on medical equipment as a vector to calculate the similarity between users, determining the user with the similarity meeting preset conditions with a target user as an adjacent neighbor user, and recommending the preference of the adjacent neighbor user to the target user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111031007.6A CN115757982A (en) | 2021-09-03 | 2021-09-03 | Medical instrument recommendation method and system based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111031007.6A CN115757982A (en) | 2021-09-03 | 2021-09-03 | Medical instrument recommendation method and system based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115757982A true CN115757982A (en) | 2023-03-07 |
Family
ID=85332397
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111031007.6A Pending CN115757982A (en) | 2021-09-03 | 2021-09-03 | Medical instrument recommendation method and system based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115757982A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116313060A (en) * | 2023-04-26 | 2023-06-23 | 海南子午互联网医院有限公司 | Digital marketing management system for Internet hospital |
| CN116705262A (en) * | 2023-06-14 | 2023-09-05 | 苏州阿基米德网络科技有限公司 | Medical equipment management and recommendation method |
| CN116894125A (en) * | 2023-09-11 | 2023-10-17 | 江苏优创生物医学科技有限公司 | Medical instrument recommendation method and system based on artificial intelligence |
-
2021
- 2021-09-03 CN CN202111031007.6A patent/CN115757982A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116313060A (en) * | 2023-04-26 | 2023-06-23 | 海南子午互联网医院有限公司 | Digital marketing management system for Internet hospital |
| CN116705262A (en) * | 2023-06-14 | 2023-09-05 | 苏州阿基米德网络科技有限公司 | Medical equipment management and recommendation method |
| CN116705262B (en) * | 2023-06-14 | 2024-04-12 | 苏州阿基米德网络科技有限公司 | Medical equipment management and recommendation method |
| CN116894125A (en) * | 2023-09-11 | 2023-10-17 | 江苏优创生物医学科技有限公司 | Medical instrument recommendation method and system based on artificial intelligence |
| CN116894125B (en) * | 2023-09-11 | 2023-11-21 | 江苏优创生物医学科技有限公司 | Medical instrument recommendation method and system based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115757982A (en) | Medical instrument recommendation method and system based on artificial intelligence | |
| Xu et al. | Applied data mining | |
| US8738534B2 (en) | Method for providing with a score an object, and decision-support system | |
| Chawla | A novel approach of cluster based optimal ranking of clicked URLs using genetic algorithm for effective personalized web search | |
| Azevedo et al. | Hybrid approaches to optimization and machine learning methods: a systematic literature review | |
| Sultanov et al. | Application of swarm techniques to requirements tracing | |
| Sharma et al. | Supervised machine learning method for ontology-based financial decisions in the stock market | |
| CN113239071B (en) | Retrieval query method and system for scientific and technological resource subject and research topic information | |
| Mladenova et al. | Analysis of the KNN classifier distance metrics for Bulgarian fake news detection | |
| Li et al. | Explain graph neural networks to understand weighted graph features in node classification | |
| CN116739794B (en) | User personalized scheme recommendation method and system based on big data and machine learning | |
| Zaffar et al. | A hybrid feature selection framework for predicting students performance | |
| Akhiat et al. | IDS-EFS: Ensemble feature selection-based method for intrusion detection system | |
| Hancer | An improved evolutionary wrapper-filter feature selection approach with a new initialisation scheme | |
| Alhenawi et al. | Improved intelligent water drop-based hybrid feature selection method for microarray data processing | |
| Aydoğdu | A new student modeling technique with convolutional neural networks: Learnerprints | |
| Xu et al. | Knowledge Discovery and Data Mining. | |
| Janani et al. | Minkowski distance measure in fuzzy PROMETHEE for ensemble feature selection | |
| Varshney et al. | Review of various Artificial Intelligence Techniques and its applications | |
| Corral et al. | Analysis of vulnerability assessment results based on CAOS | |
| Wong et al. | Intelligent data mining and personalisation for customer relationship management | |
| Déneckère et al. | A framework for comparative analysis of intention mining approaches | |
| CN114647738A (en) | Abnormal behavior correlation analysis method and system for medical insurance data | |
| Gopalan et al. | Discovering Accurate and Interesting Classification Rules Using Genetic Algorithm. | |
| Chen et al. | Application of a 3NN+ 1 based CBR system to segmentation of the notebook computers market |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |