CN113344480B - Replacement material recommendation method and system - Google Patents
Replacement material recommendation method and system Download PDFInfo
- Publication number
- CN113344480B CN113344480B CN202110905947.7A CN202110905947A CN113344480B CN 113344480 B CN113344480 B CN 113344480B CN 202110905947 A CN202110905947 A CN 202110905947A CN 113344480 B CN113344480 B CN 113344480B
- Authority
- CN
- China
- Prior art keywords
- replacement
- attribute
- replaced
- acquiring
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/23—Updating
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Artificial Intelligence (AREA)
- Strategic Management (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Development Economics (AREA)
- Computing Systems (AREA)
- Educational Administration (AREA)
- Software Systems (AREA)
- Game Theory and Decision Science (AREA)
- Evolutionary Computation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a replacement material recommendation method and a system, wherein the method comprises the following steps: acquiring material replacement demand information, wherein the material replacement demand information at least comprises one of the following information: the name of the material to be replaced and the attribute information of the material to be replaced; acquiring a final replacement condition according to the name of the material to be replaced and/or the attribute information of the material to be replaced, wherein the final replacement condition comprises the following steps: one or more target attribute tags; inputting the final replacement condition into a replacement material recommendation model for replacing material recommendation, and acquiring a replacement material sequence, wherein the replacement material sequence comprises: a plurality of replacement materials; according to the replacement materials, acquiring purchase information corresponding to the replacement materials, wherein the purchase information comprises: purchase route, purchase price, delivery location and estimated time of arrival; sequencing a plurality of replacement materials in the replacement material sequence according to the purchase information and a preset sequencing rule to complete replacement material recommendation; the method realizes intelligent recommendation of the replacement materials, and has high recommendation accuracy.
Description
Technical Field
The invention relates to the field of data processing, in particular to a replacement material recommendation method and system.
Background
During enterprise development or experimentation, replacement materials are often required, such as: the materials with the same function and lower price are used as the replacement materials. At present, generally, according to the characteristics of waiting to replace the material, adopt artificial mode, select the replacement material that accords with the demand, it is lower nevertheless to adopt artificial mode to look for replacement material efficiency, and because people's knowledge is limited, but the replaceability that leads to the replacement material of looking for easily and original material is lower, brings very big inconvenience for enterprise and user.
Disclosure of Invention
The invention provides a replacement material recommendation method and system, and aims to solve the problems that in the prior art, the efficiency of finding replacement materials in a manual mode is low, and the replaceability of the found replacement materials and original materials is low.
The invention provides a replacement material recommendation method, which comprises the following steps:
acquiring material replacement demand information, wherein the material replacement demand information at least comprises one of the following information: the name of the material to be replaced and the attribute information of the material to be replaced;
obtaining a final replacement condition according to the name of the material to be replaced and/or the attribute information of the material to be replaced, wherein the final replacement condition comprises: one or more target attribute tags;
inputting the final replacement condition into a replacement material recommendation model for replacement material recommendation, and acquiring a replacement material sequence, wherein the replacement material sequence comprises: a plurality of replacement materials;
according to the replacement materials, acquiring purchase information corresponding to the replacement materials, wherein the purchase information comprises: purchase route, purchase price, delivery location and estimated time of arrival;
sequencing the plurality of replacement materials according to the purchase information and a preset sequencing rule to complete replacement material recommendation;
the step of obtaining the final replacement condition according to the name of the material to be replaced and/or the attribute information of the material to be replaced comprises the following steps:
inputting the name of the material to be replaced into a preset replacement condition database for replacement condition matching, and acquiring one or more replacement attribute tags corresponding to the name of the material to be replaced, wherein the replacement condition database comprises: the system comprises a plurality of material names, a plurality of replacement attribute tags and the association degree between the replacement attribute tags and the material names, wherein the replacement attribute tags correspond to the material names;
acquiring a first replacement condition according to a replacement attribute label corresponding to the name of the material to be replaced and the association degree between the corresponding replacement attribute label and the name of the material to be replaced;
inputting the attribute information of the material to be replaced into a pre-trained semantic recognition model for semantic recognition, and acquiring one or more semantic labels corresponding to the attribute information of the material to be replaced;
acquiring a second replacement condition according to the semantic label;
generating a final replacement condition according to the first replacement condition and/or the second replacement condition;
the method for acquiring the replacement material recommendation model comprises the following steps:
obtaining a second set of samples, the second set of samples comprising: the system comprises a plurality of sample attribute labels, attribute label weight labels and real replacement material labels, wherein the sample attribute labels are correlated with the attribute label weights, and the sample attribute labels correspond to the real replacement material labels;
inputting a plurality of sample attribute labels in the second sample set into a deep neural network for feature extraction to obtain a feature vector;
obtaining a plurality of predicted replacement materials according to the feature vectors and the corresponding attribute label weights;
and training the deep neural network according to the predicted replacement material, the real replacement material label and a preset second loss function to obtain a replacement material recommendation model.
Optionally, the step of obtaining the first replacement condition includes:
according to a preset target attribute label quantity threshold value and the association degree of a plurality of replacement attribute labels and the material to be replaced, sequencing and screening the replacement attribute labels corresponding to the name of the material to be replaced to obtain a first target attribute label;
determining a first weight of a plurality of first target attribute tags according to the association degree of the first target attribute tags and the material to be replaced;
and acquiring a first replacement condition according to the first target attribute label and the corresponding first weight.
Optionally, the step of obtaining a second replacement condition according to the semantic tag includes:
acquiring the similarity between the semantic label and the attribute information of the material to be replaced;
sorting and screening the semantic tags according to the similarity and a preset target attribute tag quantity threshold value to obtain a second target attribute tag;
determining a second weight of the second target attribute label according to the similarity;
and acquiring a second replacement condition according to the second target attribute label and the corresponding second weight.
Optionally, the obtaining of the semantic recognition model includes:
obtaining a first set of samples, the first set of samples comprising: material attribute sample information and real semantic labels corresponding to the material attribute sample information;
inputting the material attribute sample information in the first sample set into a long-term and short-term memory network for semantic feature extraction, and acquiring sample semantic features;
acquiring a corresponding prediction semantic label according to the sample semantic features;
and performing iterative training on the long-term and short-term memory network according to the predicted semantic label, the real semantic label and a preset first loss function to obtain a trained semantic recognition model.
Optionally, inputting the final replacement condition into a replacement material recommendation model to perform replacement material recommendation, and the step of obtaining a replacement material sequence includes:
inputting the final replacement condition into a replacement material recommendation model, acquiring a plurality of replacement materials output by the replacement material recommendation model, and acquiring attribute labels of the replacement materials;
comparing the target attribute labels with the attribute labels of the replacement materials, and determining the coincidence degree between the target attribute labels and the attribute labels of the replacement materials;
and sequencing the plurality of replacement materials according to the coincidence degree to obtain a replacement material sequence.
Optionally, the step of sorting the plurality of replacement materials in the replacement material sequence according to the purchase information and a preset sorting rule includes:
respectively setting corresponding weights for a purchase path, a purchase price, a delivery place and predicted arrival time in the purchase information;
according to a preset scoring strategy and the weight, scoring the purchase information to obtain purchase scores of a plurality of replacement materials;
and sequencing the plurality of replacement materials in the replacement material sequence according to the purchase scores and the sequencing rules.
Optionally, sequencing a plurality of replacement materials to obtain a final material recommendation sequence;
and transmitting the final material recommendation sequence to a related terminal for visual display.
The invention also provides a replacement material recommendation system, comprising:
the replacing condition obtaining module is used for obtaining material replacing demand information, and the material replacing demand information at least comprises one of the following information: the name of the material to be replaced and the attribute information of the material to be replaced; obtaining a final replacement condition according to the name of the material to be replaced and/or the attribute information of the material to be replaced, wherein the final replacement condition comprises: one or more target attribute tags;
a replacement material sequence obtaining module, configured to input the final replacement condition into a replacement material recommendation model to perform replacement material recommendation, and obtain a replacement material sequence, where the replacement material sequence includes: a plurality of replacement materials;
the replacement material recommending module is used for acquiring purchase information corresponding to the replacement material according to the replacement material, and the purchase information comprises: purchase route, purchase price, delivery location and estimated time of arrival; sequencing a plurality of replacement materials in the replacement material sequence according to the purchase information and a preset sequencing rule, and recommending the replacement materials; the replacement condition acquisition module, the replacement material sequence acquisition module and the replacement material recommendation module are connected;
the step of obtaining the final replacement condition according to the name of the material to be replaced and/or the attribute information of the material to be replaced comprises the following steps:
inputting the name of the material to be replaced into a preset replacement condition database for replacement condition matching, and acquiring one or more replacement attribute tags corresponding to the name of the material to be replaced, wherein the replacement condition database comprises: the system comprises a plurality of material names, a plurality of replacement attribute tags and the association degree between the replacement attribute tags and the material names, wherein the replacement attribute tags correspond to the material names;
acquiring a first replacement condition according to a replacement attribute label corresponding to the name of the material to be replaced and the association degree between the corresponding replacement attribute label and the name of the material to be replaced;
inputting the attribute information of the material to be replaced into a pre-trained semantic recognition model for semantic recognition, and acquiring one or more semantic labels corresponding to the attribute information of the material to be replaced;
acquiring a second replacement condition according to the semantic label;
generating a final replacement condition according to the first replacement condition and/or the second replacement condition;
the method for acquiring the replacement material recommendation model comprises the following steps:
obtaining a second set of samples, the second set of samples comprising: the system comprises a plurality of sample attribute labels, attribute label weight labels and real replacement material labels, wherein the sample attribute labels are correlated with the attribute label weights, and the sample attribute labels correspond to the real replacement material labels;
inputting a plurality of sample attribute labels in the second sample set into a deep neural network for feature extraction to obtain a feature vector;
obtaining a plurality of predicted replacement materials according to the feature vectors and the corresponding attribute label weights;
and training the deep neural network according to the predicted replacement material, the real replacement material label and a preset second loss function to obtain a replacement material recommendation model.
The invention has the beneficial effects that: according to the replacement material recommendation method, the final replacement condition is obtained according to the name of the material to be replaced and/or the attribute information of the material to be replaced, the final replacement condition is input into a replacement material recommendation model for replacing material recommendation, and a replacement material sequence is obtained and comprises the following steps: the replacement materials are arranged, the purchase information corresponding to the replacement materials is acquired according to the replacement materials, the replacement materials in the replacement material sequence are arranged according to the purchase information and the preset arrangement rule, the replacement materials are recommended, intelligent recommendation of the replacement materials is achieved, recommendation accuracy is high, automation degree is high, recommendation efficiency is high, and labor cost is effectively reduced.
Drawings
Fig. 1 is a schematic flow chart of a replacement material recommendation method in an embodiment of the present invention.
Fig. 2 is a schematic flow chart illustrating the process of obtaining the final replacement condition in the replacement material recommendation method according to the embodiment of the present invention.
Fig. 3 is a schematic flow chart illustrating the process of obtaining the first replacement condition in the replacement material recommendation method according to the embodiment of the present invention.
Fig. 4 is a schematic flow chart illustrating the process of obtaining the second replacement condition in the replacement material recommendation method according to the embodiment of the present invention.
Fig. 5 is a schematic flow chart illustrating a semantic recognition model obtained in the replacement material recommendation method according to the embodiment of the present invention.
Fig. 6 is a schematic flow chart illustrating a process of obtaining a replacement material recommendation model in the replacement material recommendation method according to the embodiment of the present invention.
Fig. 7 is a schematic flow chart illustrating a process of acquiring a replacement material sequence in the replacement material recommendation method according to the embodiment of the present invention.
Fig. 8 is a schematic structural diagram of an alternative material recommendation system in an embodiment of the present invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the components related to the present invention are only shown in the drawings rather than drawn according to the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.
The inventor finds that during the process of enterprise development or experiment, replacement materials are generally needed, such as: the materials with the same function and lower price are used as the replacement materials. At present, generally, according to the characteristics of waiting to replace the material, adopt artificial mode, select the replacement material that accords with the demand, it is lower nevertheless to adopt artificial mode to look for replacement material efficiency, and because people's knowledge is limited, but the replaceability that leads to the replacement material of looking for easily and original material is lower, brings very big inconvenience for enterprise and user, and experience is felt relatively poorly. Therefore, the inventor proposes a replacement material recommendation method and system, a final replacement condition is obtained according to the name of the material to be replaced and/or the attribute information of the material to be replaced, the final replacement condition is input into a replacement material recommendation model for replacing material recommendation, and a replacement material sequence is obtained, wherein the replacement material sequence comprises: the method comprises the steps of replacing materials, obtaining purchase information corresponding to the replacing materials according to the replacing materials, sequencing the replacing materials in a replacing material sequence according to the purchase information and a preset sequencing rule, recommending the replacing materials, achieving intelligent recommendation of the replacing materials, and being high in recommendation accuracy, high in automation degree, high in recommendation efficiency, capable of effectively reducing labor cost and high in implementability.
As shown in fig. 1, the method for recommending replacement materials in this embodiment includes:
s101: acquiring material replacement demand information, wherein the material replacement demand information at least comprises one of the following information: the name of the material to be replaced and the attribute information of the material to be replaced; for example: in the process of medical research or experiment, when the materials need to be replaced, the names of the corresponding materials to be replaced and/or the attribute information of the materials to be replaced can be acquired. It is understood that the attribute information refers to the attributes of the material, such as material quality, hardness, and composition.
S102: obtaining a final replacement condition according to the name of the material to be replaced and/or the attribute information of the material to be replaced, wherein the final replacement condition comprises: one or more target attribute tags; the final replacement condition is obtained by utilizing the name of the material to be replaced and/or the attribute information of the material to be replaced, so that the accuracy of subsequently recommending the replacement material can be improved.
S103: inputting the final replacement condition into a replacement material recommendation model for replacement material recommendation, and acquiring a replacement material sequence, wherein the replacement material sequence comprises: a plurality of replacement materials; the final replacement condition is input into the replacement material recommendation model for recommending the replacement material, so that a replacement material sequence can be well obtained, and the replaceability is high.
S104: according to the replacement materials, acquiring purchase information corresponding to the replacement materials, wherein the purchase information comprises: purchase route, purchase price, delivery location and estimated time of arrival; through obtaining the purchase information corresponding to the replacement materials, the replacement material sequence can be conveniently and intelligently sequenced.
S105: and sequencing the plurality of replacement materials in the replacement material sequence according to the purchase information and a preset sequencing rule, and recommending the replacement materials. Through according to purchase information and sequencing rule, the replacement materials in the replacement material sequence are reordered, the intelligent recommendation of the replacement materials can be realized, the method is more humanized, the user requirements are met, and the efficiency is higher.
Referring to fig. 2, in order to obtain a more accurate final replacement condition, in this embodiment, the step of obtaining the final replacement condition according to the name of the material to be replaced and/or the attribute information of the material to be replaced includes:
s201: inputting the name of the material to be replaced into a preset replacement condition database for replacement condition matching, and acquiring one or more replacement attribute tags corresponding to the name of the material to be replaced, wherein the replacement condition database comprises: the system comprises a plurality of material names, a plurality of replacement attribute tags and the association degree between the replacement attribute tags and the material names, wherein the replacement attribute tags correspond to the material names.
S202: acquiring a first replacement condition according to a replacement attribute label corresponding to the name of the material to be replaced and the association degree between the corresponding replacement attribute label and the name of the material to be replaced; the first replacement condition is obtained through the steps, and the accuracy of recommending the subsequent replacement materials can be better improved.
S203: inputting the attribute information of the material to be replaced into a pre-trained semantic recognition model for semantic recognition, and acquiring one or more semantic labels corresponding to the attribute information of the material to be replaced; by performing semantic identification on the attribute information of the material to be replaced, the deep semantics of the attribute information of the material to be replaced can be conveniently acquired, and then the material to be replaced is recommended, so that the problem of low recommendation accuracy of the material to be replaced due to single description of the attribute information of the material to be replaced is avoided.
S204: acquiring a second replacement condition according to the semantic label; the second replacement condition is obtained through the steps, and the accuracy of recommendation of subsequent replacement materials is facilitated.
S205: and generating a final replacement condition according to the first replacement condition and/or the second replacement condition. When the first replacement condition and the second replacement condition exist, updating the weights corresponding to the target attribute labels in the first replacement condition and the second replacement condition according to the preset weight corresponding to the first replacement condition and the preset weight corresponding to the second replacement condition, and generating or acquiring a final replacement condition; when only the first replacement condition exists, taking the first replacement condition as a final replacement condition; when only the second replacement condition exists, the second replacement condition is taken as the final replacement condition.
Further, as shown in fig. 3, the step of obtaining the first replacement condition includes:
s301: according to a preset target attribute label quantity threshold value and the association degree of a plurality of replacement attribute labels and the material to be replaced, sequencing and screening the replacement attribute labels corresponding to the name of the material to be replaced to obtain a first target attribute label; for example: the replacement attribute tags are sequenced according to the association degree of the replacement attribute tags and the material to be replaced, the replacement attribute tags exceeding the target attribute quantity threshold are removed according to the target attribute quantity threshold, the first target attribute tag is obtained, and the situation that the number of the generated first target attribute tags is too large to cause too heavy system load is avoided.
S302: determining a first weight of a plurality of first target attribute tags according to the association degree of the first target attribute tags and the material to be replaced; namely, the weight of the first target attribute label is set according to the correlation degree of the first target attribute label and the material to be replaced, and the first weight is obtained. It can be understood that, when the degree of association between the first target attribute tag and the material to be replaced is higher, the weight of the corresponding first target attribute tag will also be increased, and details are not described here.
S303: and acquiring a first replacement condition according to the first target attribute label and the corresponding first weight. Namely, a first target attribute label and a corresponding first weight are used as a first replacement condition, and the first target attribute label and the first weight are in one-to-one correspondence.
Referring to fig. 4, in this embodiment, the step of obtaining the second replacement condition according to the semantic tag includes:
s401: acquiring the similarity between the semantic label and the attribute information of the material to be replaced;
s402: sorting and screening the semantic tags according to the similarity and a preset target attribute tag quantity threshold value to obtain a second target attribute tag;
s403: determining a second weight of the second target attribute label according to the similarity; and setting the weight of the second target attribute label according to the similarity of the semantic label and the attribute information of the material to be replaced, and acquiring the second weight. It is understood that when the similarity is larger, the corresponding second weight is also larger.
S404: and acquiring a second replacement condition according to the second target attribute label and the corresponding second weight. Namely, the second target attribute label and the corresponding second weight are used as a second replacement condition, and the second target attribute label and the second weight are in one-to-one correspondence.
As shown in fig. 5, the obtaining step of the semantic recognition model in this embodiment includes:
s501: obtaining a first set of samples, the first set of samples comprising: material attribute sample information and real semantic labels corresponding to the material attribute sample information; by collecting the first sample set, data support can be conveniently provided for subsequent training of the long-term and short-term memory network.
S502: inputting the material attribute sample information in the first sample set into a long-term and short-term memory network for semantic feature extraction, and acquiring sample semantic features;
s503: acquiring a corresponding prediction semantic label according to the sample semantic features;
s504: and performing iterative training on the long-term and short-term memory network according to the predicted semantic label, the real semantic label and a preset first loss function to obtain a trained semantic recognition model. The first loss function may be a common cross entropy loss function, an average error loss function, and the like, and is not described herein again.
Referring to fig. 6, in the present embodiment, the step of obtaining the replacement material recommendation model includes:
s601: obtaining a second set of samples, the second set of samples comprising: the system comprises a plurality of sample attribute labels, attribute label weight labels and real replacement material labels, wherein the sample attribute labels are correlated with the attribute label weights, and the sample attribute labels correspond to the real replacement material labels; by obtaining the second sample set, data support can be provided for subsequent training of the deep neural network.
S602: inputting a plurality of sample attribute labels in the second sample set into a deep neural network for feature extraction to obtain a feature vector;
s603: obtaining a plurality of predicted replacement materials according to the feature vectors and the corresponding attribute label weights; weighting the plurality of feature vectors according to the attribute label weight corresponding to the feature vectors, and further obtaining a plurality of corresponding predicted replacement materials. The predicted quantity of the replacement materials can be set according to actual requirements, and details are not repeated here.
S604: and training the deep neural network according to the predicted replacement material, the real replacement material label and a preset second loss function to obtain a replacement material recommendation model. By performing iterative training on the deep neural network, a better alternative material recommendation model can be obtained. The second loss function may be a common loss function such as a log-likelihood loss function, and is not described herein again.
As shown in fig. 7, inputting the final replacement condition into a replacement material recommendation model to perform replacement material recommendation, and the step of obtaining a replacement material sequence includes:
s701: inputting the final replacement condition into a replacement material recommendation model, acquiring a plurality of replacement materials output by the replacement material recommendation model, and acquiring attribute labels of the replacement materials; it will be appreciated that the attribute tags for replacement materials may be retrieved from a pre-set database.
S702: comparing the target attribute labels with the attribute labels of the replacement materials, and determining the coincidence degree between the target attribute labels and the attribute labels of the replacement materials; the coincidence degree is the coincidence quantity of the target attribute label and the attribute label of the replacement material.
S703: and sequencing the plurality of replacement materials according to the coincidence degree to obtain a replacement material sequence. For example: according to the size of the coincidence quantity of the target attribute labels and the attribute labels of the replacement materials, the replacement materials are sequenced, the sequence of the replacement materials is obtained, the replacement materials are intelligently sequenced, and the accuracy of the sequence of the replacement materials is improved.
In some embodiments, the step of sorting the plurality of replacement materials in the sequence of replacement materials according to the purchase information and a preset sorting rule includes:
respectively setting corresponding weights for a purchase path, a purchase price, a delivery place and predicted arrival time in the purchase information;
according to a preset scoring strategy and the weight, scoring the purchase information to obtain purchase scores of a plurality of replacement materials; the scoring policy may be set according to actual conditions, for example: setting different basic scores for different purchasing ways, setting different basic scores for different purchasing prices, setting different basic scores for different delivery places, setting different basic scores according to different estimated arrival times, weighting a plurality of basic scores in the purchasing information according to the weight, and acquiring the purchasing score of the replacement material.
And sequencing the plurality of replacement materials in the replacement material sequence according to the purchase scores and the sequencing rules. The ordering rule may be in ascending or descending order.
In some embodiments, sequencing a plurality of the replacement materials to obtain a final material recommendation sequence; and transmitting the final material recommendation sequence to a related terminal for visual display. Through transmitting final material recommendation sequence to the associated terminal, visual display is carried out, and the user can conveniently identify recommended alternative materials, so that the user experience is improved.
As shown in fig. 8, this embodiment further provides a replacement material recommendation system, including:
the replacing condition obtaining module is used for obtaining material replacing demand information, and the material replacing demand information at least comprises one of the following information: the name of the material to be replaced and the attribute information of the material to be replaced; obtaining a final replacement condition according to the name of the material to be replaced and/or the attribute information of the material to be replaced, wherein the final replacement condition comprises: one or more target attribute tags;
a replacement material sequence obtaining module, configured to input the final replacement condition into a replacement material recommendation model to perform replacement material recommendation, and obtain a replacement material sequence, where the replacement material sequence includes: a plurality of replacement materials;
the replacement material recommending module is used for acquiring purchase information corresponding to the replacement material according to the replacement material, and the purchase information comprises: purchase route, purchase price, delivery location and estimated time of arrival; sequencing a plurality of replacement materials in the replacement material sequence according to the purchase information and a preset sequencing rule, and recommending the replacement materials; the replacement condition acquisition module, the replacement material sequence acquisition module and the replacement material recommendation module are sequentially connected. In the replacement material recommendation system in this embodiment, a final replacement condition is obtained according to a name of a material to be replaced and/or attribute information of the material to be replaced, the final replacement condition is input to a replacement material recommendation model to recommend the replacement material, and a replacement material sequence is obtained, where the replacement material sequence includes: the replacement materials are arranged, the purchase information corresponding to the replacement materials is acquired according to the replacement materials, the replacement materials in the replacement material sequence are arranged according to the purchase information and the preset arrangement rule, the replacement materials are recommended, intelligent recommendation of the replacement materials is achieved, recommendation accuracy is high, automation degree is high, recommendation efficiency is high, and labor cost is effectively reduced.
In some embodiments, the step of obtaining the final replacement condition according to the name of the material to be replaced and/or the attribute information of the material to be replaced includes:
inputting the name of the material to be replaced into a preset replacement condition database for replacement condition matching, and acquiring one or more replacement attribute tags corresponding to the name of the material to be replaced, wherein the replacement condition database comprises: the system comprises a plurality of material names, a plurality of replacement attribute tags and the association degree between the replacement attribute tags and the material names, wherein the replacement attribute tags correspond to the material names;
acquiring a first replacement condition according to a replacement attribute label corresponding to the name of the material to be replaced and the association degree between the corresponding replacement attribute label and the name of the material to be replaced;
inputting the attribute information of the material to be replaced into a pre-trained semantic recognition model for semantic recognition, and acquiring one or more semantic labels corresponding to the attribute information of the material to be replaced;
acquiring a second replacement condition according to the semantic label;
and generating a final replacement condition according to the first replacement condition and/or the second replacement condition.
In some embodiments, the step of obtaining the first replacement condition comprises:
according to a preset target attribute label quantity threshold value and the association degree of a plurality of replacement attribute labels and the material to be replaced, sequencing and screening the replacement attribute labels corresponding to the name of the material to be replaced to obtain a first target attribute label;
determining a first weight of a plurality of first target attribute tags according to the association degree of the first target attribute tags and the material to be replaced;
and acquiring a first replacement condition according to the first target attribute label and the corresponding first weight.
In some embodiments, the step of obtaining a second replacement condition based on the semantic tag comprises:
acquiring the similarity between the semantic label and the attribute information of the material to be replaced;
sorting and screening the semantic tags according to the similarity and a preset target attribute tag quantity threshold value to obtain a second target attribute tag;
determining a second weight of the second target attribute label according to the similarity;
and acquiring a second replacement condition according to the second target attribute label and the corresponding second weight.
In some embodiments, the obtaining of the semantic recognition model comprises:
obtaining a first set of samples, the first set of samples comprising: material attribute sample information and real semantic labels corresponding to the material attribute sample information;
inputting the material attribute sample information in the first sample set into a long-term and short-term memory network for semantic feature extraction, and acquiring sample semantic features;
acquiring a corresponding prediction semantic label according to the sample semantic features;
and performing iterative training on the long-term and short-term memory network according to the predicted semantic label, the real semantic label and a preset first loss function to obtain a trained semantic recognition model.
In some embodiments, the step of obtaining the replacement materials recommendation model includes:
obtaining a second set of samples, the second set of samples comprising: the system comprises a plurality of sample attribute labels, attribute label weight labels and real replacement material labels, wherein the sample attribute labels are correlated with the attribute label weights, and the sample attribute labels correspond to the real replacement material labels;
inputting a plurality of sample attribute labels in the second sample set into a deep neural network for feature extraction to obtain a feature vector;
obtaining a plurality of predicted replacement materials according to the feature vectors and the corresponding attribute label weights;
and training the deep neural network according to the predicted replacement material, the real replacement material label and a preset second loss function to obtain a replacement material recommendation model.
In some embodiments, the final replacement condition is input into a replacement material recommendation model for replacement material recommendation, and the step of obtaining a replacement material sequence includes:
inputting the final replacement condition into a replacement material recommendation model, acquiring a plurality of replacement materials output by the replacement material recommendation model, and acquiring attribute labels of the replacement materials;
comparing the target attribute labels with the attribute labels of the replacement materials, and determining the coincidence degree between the target attribute labels and the attribute labels of the replacement materials;
and sequencing the plurality of replacement materials according to the coincidence degree to obtain a replacement material sequence.
In some embodiments, the step of sorting the plurality of replacement materials in the sequence of replacement materials according to the purchase information and a preset sorting rule includes:
respectively setting corresponding weights for a purchase path, a purchase price, a delivery place and predicted arrival time in the purchase information;
according to a preset scoring strategy and the weight, scoring the purchase information to obtain purchase scores of a plurality of replacement materials;
and sequencing the plurality of replacement materials in the replacement material sequence according to the purchase scores and the sequencing rules.
In some embodiments, sequencing a plurality of the replacement materials to obtain a final material recommendation sequence;
and transmitting the final material recommendation sequence to a related terminal for visual display.
The present embodiment also provides a computer-readable storage medium on which a computer program is stored, which when executed by a processor implements any of the methods in the present embodiments.
The present embodiment further provides an electronic terminal, including: a processor and a memory;
the memory is used for storing computer programs, and the processor is used for executing the computer programs stored by the memory so as to enable the terminal to execute the method in the embodiment.
The computer-readable storage medium in the present embodiment can be understood by those skilled in the art as follows: all or part of the steps for implementing the above method embodiments may be performed by hardware associated with a computer program. The aforementioned computer program may be stored in a computer readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
The electronic terminal provided by the embodiment comprises a processor, a memory, a transceiver and a communication interface, wherein the memory and the communication interface are connected with the processor and the transceiver and are used for completing mutual communication, the memory is used for storing a computer program, the communication interface is used for carrying out communication, and the processor and the transceiver are used for operating the computer program so that the electronic terminal can execute the steps of the method.
In this embodiment, the Memory may include a Random Access Memory (RAM), and may also include a non-volatile Memory (non-volatile Memory), such as at least one disk Memory.
The Processor may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the Integrated Circuit may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.
The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.
Claims (8)
1. A replacement material recommendation method, comprising:
acquiring material replacement demand information, wherein the material replacement demand information at least comprises one of the following information: the name of the material to be replaced and the attribute information of the material to be replaced;
obtaining a final replacement condition according to the name of the material to be replaced and/or the attribute information of the material to be replaced, wherein the final replacement condition comprises: one or more target attribute tags;
inputting the final replacement condition into a replacement material recommendation model for replacement material recommendation, and acquiring a replacement material sequence, wherein the replacement material sequence comprises: a plurality of replacement materials;
according to the replacement materials, acquiring purchase information corresponding to the replacement materials, wherein the purchase information comprises: purchase route, purchase price, delivery location and estimated time of arrival;
sequencing a plurality of replacement materials in the replacement material sequence according to the purchase information and a preset sequencing rule, and recommending the replacement materials;
the step of obtaining the final replacement condition according to the name of the material to be replaced and/or the attribute information of the material to be replaced comprises the following steps:
inputting the name of the material to be replaced into a preset replacement condition database for replacement condition matching, and acquiring one or more replacement attribute tags corresponding to the name of the material to be replaced, wherein the replacement condition database comprises: the system comprises a plurality of material names, a plurality of replacement attribute tags and the association degree between the replacement attribute tags and the material names, wherein the replacement attribute tags correspond to the material names;
acquiring a first replacement condition according to a replacement attribute label corresponding to the name of the material to be replaced and the association degree between the corresponding replacement attribute label and the name of the material to be replaced;
inputting the attribute information of the material to be replaced into a pre-trained semantic recognition model for semantic recognition, and acquiring one or more semantic labels corresponding to the attribute information of the material to be replaced;
acquiring a second replacement condition according to the semantic label;
generating a final replacement condition according to the first replacement condition and/or the second replacement condition;
the method for acquiring the replacement material recommendation model comprises the following steps:
obtaining a second set of samples, the second set of samples comprising: the system comprises a plurality of sample attribute labels, attribute label weight labels and real replacement material labels, wherein the sample attribute labels are correlated with the attribute label weights, and the sample attribute labels correspond to the real replacement material labels;
inputting a plurality of sample attribute labels in the second sample set into a deep neural network for feature extraction to obtain a feature vector;
obtaining a plurality of predicted replacement materials according to the feature vectors and the corresponding attribute label weights;
and training the deep neural network according to the predicted replacement material, the real replacement material label and a preset second loss function to obtain a replacement material recommendation model.
2. The replacement material recommendation method according to claim 1, wherein the step of obtaining a first replacement condition comprises:
according to a preset target attribute label quantity threshold value and the association degree of a plurality of replacement attribute labels and the material to be replaced, sequencing and screening the replacement attribute labels corresponding to the name of the material to be replaced to obtain a first target attribute label;
determining a first weight of a plurality of first target attribute tags according to the association degree of the first target attribute tags and the material to be replaced;
and acquiring a first replacement condition according to the first target attribute label and the corresponding first weight.
3. The replacement material recommendation method according to claim 1, wherein the step of obtaining a second replacement condition according to the semantic tag comprises:
acquiring the similarity between the semantic label and the attribute information of the material to be replaced;
sorting and screening the semantic tags according to the similarity and a preset target attribute tag quantity threshold value to obtain a second target attribute tag;
determining a second weight of the second target attribute label according to the similarity;
and acquiring a second replacement condition according to the second target attribute label and the corresponding second weight.
4. The replacement material recommendation method according to claim 1, wherein the obtaining of the semantic recognition model comprises:
obtaining a first set of samples, the first set of samples comprising: material attribute sample information and real semantic labels corresponding to the material attribute sample information;
inputting the material attribute sample information in the first sample set into a long-term and short-term memory network for semantic feature extraction, and acquiring sample semantic features;
acquiring a corresponding prediction semantic label according to the sample semantic features;
and performing iterative training on the long-term and short-term memory network according to the predicted semantic label, the real semantic label and a preset first loss function to obtain a trained semantic recognition model.
5. The replacement material recommendation method according to claim 1, wherein the final replacement condition is input into a replacement material recommendation model for replacement material recommendation, and the step of obtaining a replacement material sequence comprises:
inputting the final replacement condition into a replacement material recommendation model, acquiring a plurality of replacement materials output by the replacement material recommendation model, and acquiring attribute labels of the replacement materials;
comparing the target attribute labels with the attribute labels of the replacement materials, and determining the coincidence degree between the target attribute labels and the attribute labels of the replacement materials;
and sequencing the plurality of replacement materials according to the coincidence degree to obtain a replacement material sequence.
6. The replacement material recommendation method according to claim 1, wherein the step of ranking the plurality of replacement materials in the sequence of replacement materials according to the purchase information and a preset ranking rule comprises:
respectively setting corresponding weights for a purchase path, a purchase price, a delivery place and predicted arrival time in the purchase information;
according to a preset scoring strategy and the weight, scoring the purchase information to obtain purchase scores of a plurality of replacement materials;
and sequencing the plurality of replacement materials in the replacement material sequence according to the purchase scores and the sequencing rules.
7. The replacement material recommendation method according to claim 1, wherein a plurality of replacement materials are sorted to obtain a final material recommendation sequence;
and transmitting the final material recommendation sequence to a related terminal for visual display.
8. A replacement material recommendation system, comprising:
the replacing condition obtaining module is used for obtaining material replacing demand information, and the material replacing demand information at least comprises one of the following information: the name of the material to be replaced and the attribute information of the material to be replaced; obtaining a final replacement condition according to the name of the material to be replaced and/or the attribute information of the material to be replaced, wherein the final replacement condition comprises: one or more target attribute tags;
a replacement material sequence obtaining module, configured to input the final replacement condition into a replacement material recommendation model to perform replacement material recommendation, and obtain a replacement material sequence, where the replacement material sequence includes: a plurality of replacement materials;
the replacement material recommending module is used for acquiring purchase information corresponding to the replacement material according to the replacement material, and the purchase information comprises: purchase route, purchase price, delivery location and estimated time of arrival; sequencing a plurality of replacement materials in the replacement material sequence according to the purchase information and a preset sequencing rule, and recommending the replacement materials; the replacement condition acquisition module, the replacement material sequence acquisition module and the replacement material recommendation module are connected;
the step of obtaining the final replacement condition according to the name of the material to be replaced and/or the attribute information of the material to be replaced comprises the following steps:
inputting the name of the material to be replaced into a preset replacement condition database for replacement condition matching, and acquiring one or more replacement attribute tags corresponding to the name of the material to be replaced, wherein the replacement condition database comprises: the system comprises a plurality of material names, a plurality of replacement attribute tags and the association degree between the replacement attribute tags and the material names, wherein the replacement attribute tags correspond to the material names;
acquiring a first replacement condition according to a replacement attribute label corresponding to the name of the material to be replaced and the association degree between the corresponding replacement attribute label and the name of the material to be replaced;
inputting the attribute information of the material to be replaced into a pre-trained semantic recognition model for semantic recognition, and acquiring one or more semantic labels corresponding to the attribute information of the material to be replaced;
acquiring a second replacement condition according to the semantic label;
generating a final replacement condition according to the first replacement condition and/or the second replacement condition;
the method for acquiring the replacement material recommendation model comprises the following steps:
obtaining a second set of samples, the second set of samples comprising: the system comprises a plurality of sample attribute labels, attribute label weight labels and real replacement material labels, wherein the sample attribute labels are correlated with the attribute label weights, and the sample attribute labels correspond to the real replacement material labels;
inputting a plurality of sample attribute labels in the second sample set into a deep neural network for feature extraction to obtain a feature vector;
obtaining a plurality of predicted replacement materials according to the feature vectors and the corresponding attribute label weights;
and training the deep neural network according to the predicted replacement material, the real replacement material label and a preset second loss function to obtain a replacement material recommendation model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110905947.7A CN113344480B (en) | 2021-08-09 | 2021-08-09 | Replacement material recommendation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110905947.7A CN113344480B (en) | 2021-08-09 | 2021-08-09 | Replacement material recommendation method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113344480A CN113344480A (en) | 2021-09-03 |
| CN113344480B true CN113344480B (en) | 2021-12-21 |
Family
ID=77481101
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110905947.7A Active CN113344480B (en) | 2021-08-09 | 2021-08-09 | Replacement material recommendation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113344480B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112101852A (en) * | 2020-08-11 | 2020-12-18 | 大亚湾核电运营管理有限责任公司 | Method, device, equipment and medium for recommending material sending batches of nuclear power station materials |
| CN112163553A (en) * | 2020-10-15 | 2021-01-01 | 珠海格力电器股份有限公司 | Material price accounting method and device, storage medium and computer equipment |
| CN112287190A (en) * | 2020-11-03 | 2021-01-29 | 珠海格力电器股份有限公司 | Replacement material recommendation method, device and equipment |
| CN112712165A (en) * | 2020-12-30 | 2021-04-27 | 杭州电子科技大学 | Cigarette formula intelligent maintenance method based on graph convolution neural network |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7196559B2 (en) * | 2018-11-22 | 2022-12-27 | セイコーエプソン株式会社 | Electronics |
-
2021
- 2021-08-09 CN CN202110905947.7A patent/CN113344480B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112101852A (en) * | 2020-08-11 | 2020-12-18 | 大亚湾核电运营管理有限责任公司 | Method, device, equipment and medium for recommending material sending batches of nuclear power station materials |
| CN112163553A (en) * | 2020-10-15 | 2021-01-01 | 珠海格力电器股份有限公司 | Material price accounting method and device, storage medium and computer equipment |
| CN112287190A (en) * | 2020-11-03 | 2021-01-29 | 珠海格力电器股份有限公司 | Replacement material recommendation method, device and equipment |
| CN112712165A (en) * | 2020-12-30 | 2021-04-27 | 杭州电子科技大学 | Cigarette formula intelligent maintenance method based on graph convolution neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113344480A (en) | 2021-09-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112632385B (en) | Course recommendation method, course recommendation device, computer equipment and medium | |
| CN106355449B (en) | User selection method and device | |
| CN110717099B (en) | Method and terminal for recommending film | |
| CN104834651B (en) | Method and device for providing high-frequency question answers | |
| CN110503506B (en) | Item recommendation method, device and medium based on grading data | |
| CN108846097B (en) | User interest tag representation method, article recommendation device and equipment | |
| CA3059929C (en) | Text searching method, apparatus, and non-transitory computer-readable storage medium | |
| CN107291840B (en) | User attribute prediction model construction method and device | |
| CN110852360A (en) | Image emotion recognition method, device, equipment and storage medium | |
| CN112395487B (en) | Information recommendation method and device, computer readable storage medium and electronic equipment | |
| CN103761254A (en) | Method for matching and recommending service themes in various fields | |
| CN110737756B (en) | Method, apparatus, device and medium for determining answer to user input data | |
| CN110276382B (en) | Crowd classification method, device and medium based on spectral clustering | |
| CN112818218B (en) | Information recommendation method, device, terminal equipment and computer readable storage medium | |
| CN112328909B (en) | Information recommendation method and device, computer equipment and medium | |
| CN111914159B (en) | Information recommendation method and terminal | |
| US20200210776A1 (en) | Question answering method, terminal, and non-transitory computer readable storage medium | |
| CN110852785B (en) | User grading method, device and computer readable storage medium | |
| CN112699945A (en) | Data labeling method and device, storage medium and electronic device | |
| CN110598120A (en) | Behavior data based financing recommendation method, device and equipment | |
| CN105164672A (en) | Content classification | |
| CN112380421A (en) | Resume searching method and device, electronic equipment and computer storage medium | |
| CN113704620A (en) | User label updating method, device, equipment and medium based on artificial intelligence | |
| CN116383458B (en) | Information pushing method and device | |
| CN113344480B (en) | Replacement material recommendation method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220711 Address after: 201615 room 1904, G60 Kechuang building, No. 650, Xinzhuan Road, Songjiang District, Shanghai Patentee after: Shanghai Mingping Medical Data Technology Co.,Ltd. Address before: 102400 no.86-n3557, Wanxing Road, Changyang, Fangshan District, Beijing Patentee before: Mingpinyun (Beijing) data Technology Co.,Ltd. |
|
| TR01 | Transfer of patent right |