CN112000854A - Cross-modal recommendation method and device oriented to essential computing and reasoning fusion - Google Patents
Cross-modal recommendation method and device oriented to essential computing and reasoning fusion Download PDFInfo
- Publication number
- CN112000854A CN112000854A CN202010856960.3A CN202010856960A CN112000854A CN 112000854 A CN112000854 A CN 112000854A CN 202010856960 A CN202010856960 A CN 202010856960A CN 112000854 A CN112000854 A CN 112000854A
- Authority
- CN
- China
- Prior art keywords
- resource
- conversion
- cost value
- resource group
- resources
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a cross-modal recommendation method, a cross-modal recommendation device and a cross-modal recommendation medium for essential computing and reasoning fusion, wherein a pre-stored resource information base is utilized to construct a data map, an information spectrogram and a knowledge map for an acquired target task; and matching the acquired user resource information with the data map, the information spectrogram and the knowledge map to determine the difference resources. And calculating the conversion cost value of the difference resources for resource point conversion according to a pre-established resource fusion cost library. Calculating the access cost value corresponding to each conversion resource group according to the access cost value and the access times corresponding to each resource type; and taking the sum of the access cost value and the conversion cost value corresponding to each conversion resource group as the final cost value of each conversion resource group, and selecting the conversion resource group with the minimum cost value as the recommended resource. By dynamically adjusting the difference resources and combining the access cost value of the user to the resources, the resource recommendation effect is effectively improved.
Description
Technical Field
The invention relates to the technical field of data resource analysis, in particular to a cross-modal recommendation method and device oriented to essential computing and reasoning fusion and a computer readable storage medium.
Background
The data, information and knowledge are basic resources in social production activities and can be represented by multimedia such as numbers, characters, symbols, graphs, sounds, movies and the like. Wherein the data is an original record obtained by perceiving objective things according to some measure. The information is produced by custom processing according to a certain development stage and its purpose. Knowledge is the result of systematic refinement, study, and analysis of information acquired or accumulated by knowledge workers using the brain.
In the prior art, resources are often analyzed and processed independently according to three types of data resources, information resources and knowledge resources, and the fusion among the resources is poor, so that the effect of resource recommendation depending on the data resources, the information resources and the knowledge resources is not ideal.
Therefore, how to improve the effect of resource recommendation is a problem to be solved by those skilled in the art.
Disclosure of Invention
The embodiment of the invention aims to provide a cross-modal recommendation method and device oriented to essential computing and reasoning fusion and a computer readable storage medium, which can improve the effect of resource recommendation.
In order to solve the above technical problem, an embodiment of the present invention provides a cross-modal recommendation method oriented to essential computing and inference fusion, including:
constructing a data map, an information spectrogram and a knowledge map for the acquired target task by utilizing a pre-stored resource information base;
matching the acquired user resource information with the data map, the information spectrogram and the knowledge map to determine difference resources;
calculating the conversion cost value of the difference resources for resource point conversion according to a pre-established resource fusion cost library; the resource fusion cost library comprises conversion cost values for converting data resources, information resources and knowledge resources; each pair of the difference resources performs the conversion of the resource point once to obtain a conversion resource group, and each conversion resource group has a corresponding conversion cost value;
calculating the access cost value corresponding to each conversion resource group according to the access cost value and the access times corresponding to each resource type;
taking the sum of the access cost value and the conversion cost value corresponding to each conversion resource group as the final cost value of each conversion resource group;
and selecting the conversion resource group with the minimum cost value as a recommended resource.
Optionally, the calculating, according to a pre-established resource fusion cost library, a conversion cost value of the difference resource for resource point conversion includes:
setting labels for all resource points contained in the difference resources according to the resource types of the target tasks and the resource types of all resource points in the difference resources; wherein the tags include switchable tags and non-switchable tags;
dividing the resource points provided with the convertible labels into a first resource group; dividing the resource points provided with the non-convertible labels into a second resource group;
performing resource transformation on resource points in the first resource group to obtain a plurality of initial transformation resource groups; wherein, an initial transformation resource group is obtained by executing the transformation of the resource point each time;
sequentially inquiring initial conversion cost values corresponding to all resource points in each initial conversion resource group from the resource fusion cost library to obtain a first conversion cost value corresponding to each initial conversion resource group; judging whether a target conversion resource group with a first conversion cost value smaller than the directly added cost value corresponding to the first resource group exists or not;
when a target conversion resource group with a first conversion cost value smaller than the directly added cost value corresponding to the first resource group exists, combining the target conversion resource group and the second resource group to serve as a conversion resource group, and taking the sum of the first conversion cost value corresponding to the target conversion resource group and the second cost value corresponding to the second resource group as the conversion cost value of the conversion resource group;
and when a target conversion resource group with a first conversion cost value smaller than the directly added cost value corresponding to the first resource group does not exist, combining the first resource group and the second resource group to serve as a conversion resource group, and taking the sum of the directly added cost value corresponding to the first resource group and the second cost value corresponding to the second resource group as the conversion cost value of the conversion resource group.
Optionally, sequentially querying the initial conversion cost values corresponding to all resource points in each initial conversion resource group from the resource fusion cost library to obtain the first conversion cost value corresponding to each initial conversion resource group includes:
dividing resource points for executing resource conversion in the initial conversion resource group into first type resource points; dividing resource points which do not execute resource conversion in the initial conversion resource group into second type resource points;
inquiring the direct new cost value corresponding to the second type of resource points from the resource fusion cost library;
inquiring the initial conversion cost value corresponding to the first type of resource point from the resource fusion cost library according to the resource type before the first type of resource point performs resource conversion and the resource type after the resource conversion is performed; wherein the initial conversion cost value comprises a temporal cost value, a spatial consumption cost value, a bandwidth consumption cost value, and a composite cost value;
and taking the sum of the directly added cost value and the initial conversion cost value as the first conversion cost value of the initial conversion resource group.
Optionally, the matching the acquired user resource information with the data map, the information spectrogram and the knowledge map, and determining the difference resource includes:
converting the resource information of the user into a user data map, a user information map and a user knowledge map;
matching the user data map with the data map to obtain data difference resources; matching the user information map with the information map to obtain information difference resources; matching the user knowledge graph with the knowledge graph to obtain knowledge difference resources;
and summarizing the data difference resources, the information difference resources and the knowledge difference resources to be used as difference resources.
Optionally, the calculating, according to the access cost value and the access frequency corresponding to each resource type, the access cost value corresponding to each conversion resource group includes:
determining data access cost values corresponding to all data resources in the conversion resource group according to a preset unit data volume access cost value and the total data amount contained in the conversion resource group;
determining information access cost values corresponding to all information resources in the conversion resource group according to the information access cost values corresponding to all path organization modes;
determining knowledge access cost values corresponding to all knowledge resources in the conversion resource group according to preset single access cost values of various knowledge rules and the types and the number of the knowledge rules contained in the conversion resource group;
and taking the sum of the data access cost value, the information access cost value and the knowledge access cost value as the access cost value of the conversion resource group.
Optionally, the constructing a data map, an information spectrogram and a knowledge map of the acquired target task by using a pre-stored resource information base includes:
decomposing the acquired target task according to a preset classification rule to obtain a plurality of subtasks; each subtask consists of a plurality of track nodes;
obtaining an influence value of the target subtask according to the positive value, the negative value and the required value corresponding to each track node in the target subtask; the target subtask is any one of all the subtasks;
selecting and sequencing the influence values of the subtasks by using a directed graph longest path selection algorithm to obtain an optimal path;
inquiring a pre-stored resource information base to determine resources corresponding to each track node contained in the optimal path; wherein the resources comprise data resources, information resources and knowledge resources;
constructing a data map according to all the data resources; constructing an information map according to all the information resources; and constructing a knowledge graph according to all the knowledge resources.
The embodiment of the invention also provides a cross-modal recommendation device oriented to essential computing and reasoning fusion, which comprises a construction unit, a determination unit, a conversion computing unit, an access computing unit, a summary unit and a selection unit;
the construction unit is used for constructing a data map, an information spectrogram and a knowledge map for the acquired target task by utilizing a pre-stored resource information base;
the determining unit is configured to match the acquired user resource information with the data map, the information spectrogram and the knowledge map, and determine a difference resource;
the conversion calculation unit is used for calculating the conversion cost value of the difference resource for resource point conversion according to a pre-established resource fusion cost library; the resource fusion cost library comprises conversion cost values for converting data resources, information resources and knowledge resources; each pair of the difference resources performs the conversion of the resource point once to obtain a conversion resource group, and each conversion resource group has a corresponding conversion cost value;
the access calculation unit is used for calculating the access cost value corresponding to each conversion resource group according to the access cost value and the access frequency corresponding to each resource type;
the summarizing unit is used for taking the sum of the access cost value and the conversion cost value corresponding to each conversion resource group as the final cost value of each conversion resource group;
and the selecting unit is used for selecting the conversion resource group with the minimum cost value as the recommended resource.
Optionally, the computing unit includes a setting subunit, a dividing subunit, a transforming subunit, a querying subunit, a judging subunit, a first serving subunit and a second serving subunit;
the setting subunit is configured to set a label for each resource point included in the difference resource according to the resource type of the target task and the resource type of each resource point in the difference resource; wherein the tags include switchable tags and non-switchable tags;
the dividing subunit is used for dividing the resource points provided with the convertible labels into a first resource group; dividing the resource points provided with the non-convertible labels into a second resource group;
the transformation sub-unit is used for transforming the resources of the resource points in the first resource group to obtain a plurality of initial transformation resource groups; wherein, an initial transformation resource group is obtained by executing the transformation of the resource point each time;
the query subunit is configured to sequentially query, from the resource fusion cost library, initial conversion cost values corresponding to all resource points in each initial conversion resource group to obtain a first conversion cost value corresponding to each initial conversion resource group;
the judging subunit is configured to judge whether there is a target conversion resource group of which a first conversion cost value is smaller than a directly added cost value corresponding to the first resource group;
the first serving subunit is configured to, when there is a target conversion resource group with a first conversion cost value smaller than a directly added cost value corresponding to the first resource group, combine the target conversion resource group and the second resource group to serve as a conversion resource group, and use a sum of a first conversion cost value corresponding to the target conversion resource group and a second cost value corresponding to the second resource group as a conversion cost value of the conversion resource group;
and the second serving as a subunit, configured to, when there is no target conversion resource group in which the first conversion cost value is smaller than the directly added cost value corresponding to the first resource group, combine the first resource group and the second resource group to serve as a conversion resource group, and use a sum of the directly added cost value corresponding to the first resource group and the second cost value corresponding to the second resource group as the conversion cost value of the conversion resource group.
Optionally, the query subunit is specifically configured to divide resource points, which execute resource conversion in the initial conversion resource group, into first class resource points; dividing resource points which do not execute resource conversion in the initial conversion resource group into second type resource points; inquiring the direct new cost value corresponding to the second type of resource points from the resource fusion cost library; inquiring the initial conversion cost value corresponding to the first type of resource point from the resource fusion cost library according to the resource type before the first type of resource point performs resource conversion and the resource type after the resource conversion is performed; wherein the initial conversion cost value comprises a temporal cost value, a spatial consumption cost value, a bandwidth consumption cost value, and a composite cost value; and taking the sum of the directly added cost value and the initial conversion cost value as the first conversion cost value of the initial conversion resource group.
Optionally, the determining unit comprises a transforming subunit, a matching subunit and a summarizing subunit;
the conversion unit is used for converting the resource information of the user into a user data map, a user information map and a user knowledge map;
the matching subunit is configured to match the user data map with the data map to obtain a data difference resource; matching the user information map with the information map to obtain information difference resources; matching the user knowledge graph with the knowledge graph to obtain knowledge difference resources;
the summarizing subunit is configured to summarize the data difference resource, the information difference resource, and the knowledge difference resource as difference resources.
Optionally, the access calculation unit comprises a data access subunit, an information access subunit, a knowledge access subunit, and a summation subunit;
the data access subunit is configured to determine, according to a preset unit data volume access cost value and a total amount of data included in the conversion resource group, data access cost values corresponding to all data resources in the conversion resource group;
the information access subunit is used for determining the information access cost values corresponding to all the information resources in the conversion resource group according to the information access cost values corresponding to the various path organization modes;
the knowledge access subunit is used for determining knowledge access cost values corresponding to all knowledge resources in the conversion resource group according to preset single access cost values of various knowledge rules and the types and the number of the knowledge rules contained in the conversion resource group;
the sum subunit is configured to use the sum of the data access cost value, the information access cost value, and the knowledge access cost value as the access cost value of the conversion resource group.
Optionally, the building unit includes a decomposition subunit, an obtaining subunit, a sorting subunit, a determining subunit, and a building subunit;
decomposing the acquired target task according to a preset classification rule to obtain a plurality of subtasks; each subtask consists of a plurality of track nodes;
obtaining an influence value of the target subtask according to the positive value, the negative value and the required value corresponding to each track node in the target subtask; the target subtask is any one of all the subtasks;
selecting and sequencing the influence values of the subtasks by using a directed graph longest path selection algorithm to obtain an optimal path;
inquiring a pre-stored resource information base to determine resources corresponding to each track node contained in the optimal path; wherein the resources comprise data resources, information resources and knowledge resources;
constructing a data map according to all the data resources; constructing an information map according to all the information resources; and constructing a knowledge graph according to all the knowledge resources.
The embodiment of the invention also provides a cross-modal recommendation device for essential computing and reasoning fusion, which comprises:
a memory for storing a computer program;
a processor for executing the computer program to implement the steps of the cross-modality recommendation method oriented to essential computing and reasoning fusion as described in any one of the above.
An embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the cross-modality recommendation method oriented to essential computing and inference fusion are implemented as any one of the above.
According to the technical scheme, the data map, the information spectrogram and the knowledge map are constructed for the acquired target task by utilizing the pre-stored resource information base; and matching the acquired user resource information with the data map, the information spectrogram and the knowledge map to determine the difference resources. The difference resource is used to represent the resource that needs to be provided to the user. Considering that mutual transformation and fusion reasoning can be carried out among the data map, the information spectrogram and the knowledge map, in order to recommend resources to users more reasonably, the transformation cost value of resource point transformation of the difference resources can be calculated according to a pre-established resource fusion cost library; the resource fusion cost library comprises conversion cost values for converting data resources, information resources and knowledge resources; and executing the conversion of the resource point once for each pair of different resources to obtain a conversion resource group, wherein each conversion resource group has a corresponding conversion cost value. The access cost of the user to different resource types is different, and the access cost value corresponding to each conversion resource group can be calculated according to the access cost value and the access times corresponding to each resource type; and taking the sum of the access cost value and the conversion cost value corresponding to each conversion resource group as the final cost value of each conversion resource group, and selecting the conversion resource group with the minimum cost value as the recommended resource. By converting the data resources, the information resources and the knowledge resources of the resource points contained in the difference resources, the dynamic adjustment of the difference resources can be realized, and the conversion resource group with the minimum cost value can be recommended to the user by combining the access cost value of the user to the resources, so that the resource recommendation effect is effectively improved.
Drawings
In order to illustrate the embodiments of the present invention more clearly, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings can be obtained by those skilled in the art without inventive effort.
Fig. 1 is a flowchart of a cross-modal recommendation method oriented to essential computing and inference fusion according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method for computing cost values for resources according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a cross-modal recommendation apparatus oriented to essential computing and inference fusion according to an embodiment of the present invention;
fig. 4 is a schematic hardware structure diagram of a cross-modal recommendation apparatus oriented to essential computing and inference fusion according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings 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 of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without any creative work belong to the protection scope of the present invention.
In order that those skilled in the art will better understand the disclosure, the invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
Next, a cross-modal recommendation method oriented to essential computing and inference fusion provided by the embodiment of the present invention is described in detail. Fig. 1 is a flowchart of a cross-modality recommendation method oriented to essential computing and inference fusion, provided in an embodiment of the present invention, where the method includes:
s101: and constructing a data map, an information spectrogram and a knowledge map for the acquired target task by utilizing a pre-stored resource information base.
The target task is different in form according to different application scenes. For example, the target task may be to examine lawyer qualifications, and the resource recommended to the user may be a learning plan that the user needs to perform in order to successfully examine lawyer qualifications. The learning plan includes knowledge points that the user needs to learn, and each knowledge point can be regarded as a resource point.
In the embodiment of the invention, the resources corresponding to the target task are analyzed and processed from three aspects of data, information and knowledge. Data, information, and knowledge can be viewed as three different modalities of resources.
For convenience of description, in the embodiments of the present invention, the description is given by taking an example of recommending a learning plan to a user. In the embodiment of the invention, the resources corresponding to the target task can be counted in a big data analysis mode depending on historical data. The resource corresponding to the target task is the content that the user needs to learn to achieve the target.
In a specific implementation, the acquired target task may be decomposed according to a preset classification rule to obtain a plurality of subtasks. For example, taking the lawyer qualification as the target task, the target task can be divided into a plurality of subtasks such as learning basic knowledge, learning test outline, and practicing calendar year test paper.
Wherein each subtask may be composed of a plurality of trace nodes. The trajectory nodes can be considered as knowledge points that need to be learned to complete the subtask.
The determination of the track node in each subtask is influenced by various factors, and in the embodiment of the invention, the influencing factors can be divided into three types, which are respectively represented by positive values, negative values and demand values.
The processing mode of each subtask is similar, taking any one of all subtasks, namely a target subtask as an example, the influence value of the target subtask can be obtained according to the positive value, the negative value and the required value corresponding to each track node in the target subtask; selecting and sequencing the influence values of the subtasks by using a directed graph longest path selection algorithm to obtain an optimal path; inquiring a pre-stored resource information base to determine resources corresponding to each track node contained in the optimal path; the resources comprise data resources, information resources and knowledge resources; constructing a data map according to all data resources; constructing an information map according to all information resources; and constructing a knowledge graph according to all knowledge resources.
The positive Value refers to the Value which can be provided by the track Node for completing the subtasks, can be represented by Node _ Value and takes the positive Value; the negative value refers to the Cost that a user needs to pay to achieve the subtask represented by the track Node, and can be represented by Node _ Cost, and a negative value is taken; the last type is a required value, which is represented by Node _ Other, and the value of the required value can be preset by a user according to a target task to be realized.
In the embodiment of the present invention, the Final _ Purpose may be used to represent a target task, and the target task is decomposed to obtain a plurality of subtasks, that is, Final _ Purpose1+ Purpose2+. + Purpose.
For each subtask, reaching the subtask requires a combination of multiple processes (processes). A Trace Node may be simply understood as a small step of a Process decomposition, which is denoted by Node1, Node 2., and Node n, where Trace nodes are connected by arrowed lines to form a Process Trace, and Trace is Node1+ Node 2. + Node n.
The Influence value of the target task may be added by the Influence values of the plurality of subtasks, and the formula of the Influence value of the target task (inflence _ value enode i) is as follows:
Final_Purpose＝Purpose1+Purpose2+...+PurposeN；
Purposei＝f(Trace)＝f(Process)＝f(Node1+Node2+...+NodeN)
＝f(Node1(Node_Value,Node_Cost,Node_Other)+Node2(Node_Value,Node_Cost,Node_Other)+...+NodeN(Node_Value,Node_Cost,Node_Other))；
Influence_ValueNodei(Node_Value,Node_Cost,Node_Other)＝α*g(Node_Value)+β*h(Node_Cost)+θ*p(Node_Other)；α+β+θ＝1；
the values of the α, β, and θ can be preset by a user according to a target task to be realized.
S102: and matching the acquired user resource information with the data map, the information spectrogram and the knowledge map to determine the difference resources.
The user resource information can be regarded as the resource that the user has learned to realize the target task, and the difference resource can be regarded as the resource that the user needs to learn further to realize the target task.
In the embodiment of the invention, the resources for realizing the target task are displayed in the forms of the data map, the information spectrogram and the knowledge map, so that when the difference resources are determined, the resource information of the user can be converted into the user data map, the user information map and the user knowledge map; matching the user data map with the data map to obtain data difference resources; matching the user information map with the information map to obtain information difference resources; matching the user knowledge graph with a knowledge graph to obtain knowledge difference resources; and summarizing the data difference resources, the information difference resources and the knowledge difference resources to be used as difference resources.
S103: and calculating the conversion cost value of the difference resources for resource point conversion according to a pre-established resource fusion cost library.
Resources which can perform resource conversion exist among the data resources, the information resources and the knowledge resources. For example, some data resources may be converted into information resources, some information resources may be converted into knowledge resources, and so on. In the embodiment of the invention, a resource fusion cost library can be established for the resources capable of executing resource conversion. The resource fusion cost library can include conversion cost values for conversion among data resources, information resources and knowledge resources
The number of resource points in the difference resource, which can be used for resource transformation, is often large, and in the embodiment of the present invention, a resource obtained by performing one transformation on the difference resource is referred to as a transformation resource group. And executing the conversion of the resource point once for each pair of different resources to obtain a conversion resource group, wherein each conversion resource group has a corresponding conversion cost value.
In practical application, when the number of resource points which can execute resource transformation in the different resources is less, only one resource point can be transformed at a time; when the number of resource points capable of executing resource transformation in the different resources is large, the resource points can be transformed according to the set number.
S104: and calculating the access cost value corresponding to each conversion resource group according to the access cost value and the access times corresponding to each resource type.
In embodiments of the present invention, resources are presented in the form of data, information, and knowledge. When a user acquires the learning content from the resources of data, information and knowledge forms, corresponding access cost values need to be paid, so that in the embodiment of the invention, the cost values of the resources include the access cost values in addition to the conversion cost values.
For resources of different forms, the access cost values to be paid when the user accesses the content corresponding to the resources are different.
The access cost value of each conversion resource group is calculated in a similar manner, and in the embodiment of the present invention, an example of any one conversion resource group in all conversion resource groups is described.
A set of conversion resources often includes data resources, information resources, and knowledge resources.
Taking data resources as an example, the data access cost values corresponding to all data resources in the conversion resource group can be determined according to the preset unit data volume access cost value and the total data volume contained in the conversion resource group.
Taking the information resource as an example, the information access cost values corresponding to all the information resources in the conversion resource group can be determined according to the information access cost values corresponding to the organization modes of the paths.
Taking knowledge resources as an example, the knowledge access cost values corresponding to all knowledge resources in the conversion resource group can be determined according to the preset single access cost values of various knowledge rules and the types and the number of the knowledge rules contained in the conversion resource group.
After determining the access cost values of the various types of resources in the conversion resource group, the sum of the data access cost value, the information access cost value and the knowledge access cost value can be used as the access cost value of the conversion resource group.
S105: and taking the sum of the access cost value and the conversion cost value corresponding to each conversion resource group as the final cost value of each conversion resource group.
In embodiments of the invention, resources may be evaluated in terms of both access cost values and conversion cost values. Therefore, in practical application, the sum of the access cost value and the conversion cost value corresponding to each conversion resource group can be used as the final cost value of each conversion resource group.
S106: and selecting the conversion resource group with the minimum cost value as a recommended resource.
The smaller the cost value of a resource, the more the learning plan contained in the resource fits the target task. In the embodiment of the invention, the conversion resource group with the minimum cost value can be used as the recommended resource.
In the embodiment of the invention, for the resource capable of resource conversion, the original type of the resource can be maintained, and the conversion can be performed on the resource. In practical application, the cost value of resource conversion may be compared with the cost value of the original type of resource retention, so as to determine an optimal conversion manner, as shown in fig. 2, which is a flowchart of a method for calculating a cost value of resource conversion provided in an embodiment of the present invention, the method includes:
s201: and setting labels for each resource point contained in the difference resource according to the resource type of the target task and the resource type of each resource point in the difference resource.
Wherein the label comprises a convertible label and a non-convertible label.
The resource types include three types, namely data resources, information resources and knowledge resources. In practical application, there may be a part of resources that cannot be converted into other forms of resources, so in order to facilitate subsequent conversion processing, in the embodiment of the present invention, a tag may be set for each resource point included in a difference resource, and a convertible tag may be set for a resource point that can perform resource conversion; and setting a non-convertible label for the resource point which can not execute the resource conversion.
The setting of each resource point label can be set by the user for each resource point in turn, and the resource points contained in the difference resources can be matched with the transformable resources and the non-transformable resources counted by the big data, so that the label of each resource point is determined.
S202: dividing the resource points provided with the convertible labels into a first resource group; and dividing the resource points provided with the non-convertible labels into a second resource group.
In the embodiment of the present invention, the resource point with the convertible tag in the difference resource may be divided into a first resource group, and the resource point with the non-convertible tag in the difference resource may be divided into a second resource group.
S203: and carrying out resource transformation on the resource points in the first resource group to obtain a plurality of initial transformation resource groups.
Wherein each time the resource point is converted, an initial conversion resource group can be obtained.
In practical applications, the number of resource points that need to be converted each time resource conversion is performed may be determined according to the number of resource points included in the first resource group.
S204: and sequentially inquiring the initial conversion cost values corresponding to all resource points in each initial conversion resource group from the resource fusion cost library so as to obtain the first conversion cost values corresponding to each initial conversion resource group.
In practical application, resource points for performing resource transformation in the initial transformation resource group can be divided into first type resource points; dividing resource points which do not execute resource conversion in the initial conversion resource group into second type resource points; inquiring the direct new cost value corresponding to the second type of resource points from the resource fusion cost library; and inquiring the initial conversion cost value corresponding to the first type of resource point from the resource fusion cost library according to the resource type before the first type of resource point performs resource conversion and the resource type after the resource conversion is performed. And taking the sum of the directly added cost value and the initial conversion cost value as the first conversion cost value of the initial conversion resource group.
The initial transformation Cost value may include a time Cost value (Cost _ transit time), a space consumption Cost value (Cost _ transit), a bandwidth consumption Cost value (Cost _ TransBandwidth), and a composite Cost value (Cost _ transit), among others.
The conversion cost values corresponding to different types of resources when performing conversion are different, and in the embodiment of the present invention, the conversion cost values of resource conversion performed by different resource points may be recorded in the following manner as shown in table 1.
TABLE 1
In combination with the form of conversion cost values shown in table 1, "D" represents a data resource, "I" represents an information resource, and "K" represents a knowledge resource. For example, Cost _ TransD-D represents the conversion Cost value of the data resource conversion of the same type, and Cost _ TransD-I represents the conversion Cost value of the data resource conversion into the information resource; cost _ TransD-K represents the Cost value of converting data into knowledge resources.
S205: and judging whether a target conversion resource group with the first conversion cost value smaller than the directly added cost value corresponding to the first resource group exists.
For the convertible resource, the conversion of the resource can be performed, and the resource type can be kept unchanged. In the embodiment of the invention, for the resource with the unchanged resource type, the resource can be directly recommended to the user as the newly added resource, so that the cost value of recommending the resource to the user can be represented by adopting the directly added cost value.
When a target conversion resource group with the first conversion cost value smaller than the directly added cost value corresponding to the first resource group exists, executing S206; and if the target conversion resource group with the first conversion cost value smaller than the directly added cost value corresponding to the first resource group does not exist, executing S207.
S206: and merging the target conversion resource group and the second resource group to serve as a conversion resource group, and taking the sum of the first conversion cost value corresponding to the target conversion resource group and the second generation value corresponding to the second resource group as the conversion cost value of the conversion resource group.
The second resource group refers to a resource which can not be subjected to resource conversion, and can be directly recommended to the user as a new resource. And the directly increased cost value corresponding to the second resource group is the second generation value of the second resource group.
S207: and combining the first resource group and the second resource group to serve as a conversion resource group, and taking the sum of the direct new cost value corresponding to the first resource group and the second cost value corresponding to the second resource group as the conversion cost value of the conversion resource group.
According to the technical scheme, the data map, the information spectrogram and the knowledge map are constructed for the acquired target task by utilizing the pre-stored resource information base; and matching the acquired user resource information with the data map, the information spectrogram and the knowledge map to determine the difference resources. The difference resource is used to represent the resource that needs to be provided to the user. Considering that mutual transformation and fusion reasoning can be carried out among the data map, the information spectrogram and the knowledge map, in order to recommend resources to users more reasonably, the transformation cost value of resource point transformation of the difference resources can be calculated according to a pre-established resource fusion cost library; the resource fusion cost library comprises conversion cost values for converting data resources, information resources and knowledge resources; and executing the conversion of the resource point once for each pair of different resources to obtain a conversion resource group, wherein each conversion resource group has a corresponding conversion cost value. The access cost of the user to different resource types is different, and the access cost value corresponding to each conversion resource group can be calculated according to the access cost value and the access times corresponding to each resource type; and taking the sum of the access cost value and the conversion cost value corresponding to each conversion resource group as the final cost value of each conversion resource group, and selecting the conversion resource group with the minimum cost value as the recommended resource. By converting the data resources, the information resources and the knowledge resources of the resource points contained in the difference resources, the dynamic adjustment of the difference resources can be realized, and the conversion resource group with the minimum cost value can be recommended to the user by combining the access cost value of the user to the resources, so that the resource recommendation effect is effectively improved.
Fig. 3 is a schematic structural diagram of a cross-modal recommendation apparatus for essential computing and inference fusion, provided by an embodiment of the present invention, and includes a construction unit 31, a determination unit 32, a transformation calculation unit 33, an access calculation unit 34, a summarization unit 35, and a selection unit 36;
the constructing unit 31 is configured to construct a data map, an information spectrogram and a knowledge map for the acquired target task by using a pre-stored resource information base;
a determining unit 32, configured to match the acquired user resource information with the data map, the information spectrogram, and the knowledge map, and determine a difference resource;
the conversion calculation unit 33 is configured to calculate a conversion cost value of the resource point conversion performed by the difference resource according to a pre-established resource fusion cost library; the resource fusion cost library comprises conversion cost values for converting data resources, information resources and knowledge resources; each pair of difference resources carries out the conversion of one resource point to obtain a conversion resource group, and each conversion resource group has a corresponding conversion cost value;
the access calculation unit 34 is configured to calculate access cost values corresponding to the conversion resource groups according to the access cost value and the access frequency corresponding to each resource type;
the summarizing unit 35 is used for taking the sum of the access cost value and the conversion cost value corresponding to each conversion resource group as the final cost value of each conversion resource group;
and the selecting unit 36 is configured to select the conversion resource group with the smallest cost value as the recommended resource.
Optionally, the calculation unit comprises a setting subunit, a dividing subunit, a transformation subunit, a query subunit, a judgment subunit, a first serving subunit and a second serving subunit;
the setting subunit is used for setting a label for each resource point contained in the difference resource according to the resource type of the target task and the resource type of each resource point in the difference resource; wherein the label comprises a convertible label and a non-convertible label;
a dividing subunit, configured to divide the resource points provided with the convertible labels into a first resource group; dividing the resource points provided with the non-convertible labels into a second resource group;
the transformation unit is used for carrying out resource transformation on resource points in the first resource group to obtain a plurality of initial transformation resource groups; wherein, an initial transformation resource group is obtained by executing the transformation of the resource point each time;
the query subunit is used for sequentially querying the initial conversion cost values corresponding to all the resource points in each initial conversion resource group from the resource fusion cost library so as to obtain first conversion cost values corresponding to each initial conversion resource group;
the judging subunit is used for judging whether a target conversion resource group with a first conversion cost value smaller than the directly added cost value corresponding to the first resource group exists;
the first serving subunit is used for merging the target conversion resource group and the second resource group to serve as a conversion resource group when a target conversion resource group with the first conversion cost value smaller than the directly added cost value corresponding to the first resource group exists, and taking the sum of the first conversion cost value corresponding to the target conversion resource group and the second cost value corresponding to the second resource group as the conversion cost value of the conversion resource group;
and the second as a subunit, configured to, when there is no target conversion resource group with the first conversion cost value smaller than the directly added cost value corresponding to the first resource group, combine the first resource group and the second resource group to serve as a conversion resource group, and use the sum of the directly added cost value corresponding to the first resource group and the second cost value corresponding to the second resource group as the conversion cost value of the conversion resource group.
Optionally, the query subunit is specifically configured to divide resource points, which execute resource conversion in the initial conversion resource group, into first class resource points; dividing resource points which do not execute resource conversion in the initial conversion resource group into second type resource points; inquiring the direct new cost value corresponding to the second type of resource points from the resource fusion cost library; inquiring the initial conversion cost value corresponding to the first type of resource point from the resource fusion cost library according to the resource type before the first type of resource point performs resource conversion and the resource type after performing resource conversion; wherein the initial transformation cost values comprise a time cost value, a space consumption cost value, a bandwidth consumption cost value and a comprehensive cost value; and taking the sum of the directly added cost value and the initial conversion cost value as the first conversion cost value of the initial conversion resource group.
Optionally, the determining unit comprises a transforming subunit, a matching subunit and a summarizing subunit;
the conversion unit is used for converting the resource information of the user into a user data map, a user information map and a user knowledge map;
the matching subunit is used for matching the user data map with the data map to obtain data difference resources; matching the user information map with the information map to obtain information difference resources; matching the user knowledge graph with a knowledge graph to obtain knowledge difference resources;
and the summarizing subunit is used for summarizing the data difference resource, the information difference resource and the knowledge difference resource to be used as the difference resource.
Optionally, the access calculation unit comprises a data access subunit, an information access subunit, a knowledge access subunit and a summation subunit;
the data access subunit is used for determining the data access cost values corresponding to all the data resources in the conversion resource group according to the preset unit data volume access cost value and the total data amount contained in the conversion resource group;
the information access subunit is used for determining the information access cost values corresponding to all the information resources in the conversion resource group according to the information access cost values corresponding to the path organization modes;
the knowledge access subunit is used for determining knowledge access cost values corresponding to all knowledge resources in the conversion resource group according to preset single access cost values of various knowledge rules and the types and the number of the knowledge rules contained in the conversion resource group;
and the sum subunit is used for taking the sum of the data access cost value, the information access cost value and the knowledge access cost value as the access cost value of the conversion resource group.
Optionally, the construction unit includes a decomposition subunit, an obtaining subunit, a sorting subunit, a determination subunit, and a construction subunit;
decomposing the acquired target task according to a preset classification rule to obtain a plurality of subtasks; each subtask consists of a plurality of track nodes;
obtaining the influence value of the target subtask according to the positive value, the negative value and the required value corresponding to each track node in the target subtask; the target subtask is any one of all subtasks;
selecting and sequencing the influence values of the subtasks by using a directed graph longest path selection algorithm to obtain an optimal path;
inquiring a pre-stored resource information base to determine resources corresponding to each track node contained in the optimal path; the resources comprise data resources, information resources and knowledge resources;
constructing a data map according to all data resources; constructing an information map according to all information resources; and constructing a knowledge graph according to all knowledge resources.
For the description of the features in the embodiment corresponding to fig. 3, reference may be made to the related description of the embodiments corresponding to fig. 1 and fig. 2, which is not repeated here.
According to the technical scheme, the data map, the information spectrogram and the knowledge map are constructed for the acquired target task by utilizing the pre-stored resource information base; and matching the acquired user resource information with the data map, the information spectrogram and the knowledge map to determine the difference resources. The difference resource is used to represent the resource that needs to be provided to the user. Considering that mutual transformation and fusion reasoning can be carried out among the data map, the information spectrogram and the knowledge map, in order to recommend resources to users more reasonably, the transformation cost value of resource point transformation of the difference resources can be calculated according to a pre-established resource fusion cost library; the resource fusion cost library comprises conversion cost values for converting data resources, information resources and knowledge resources; and executing the conversion of the resource point once for each pair of different resources to obtain a conversion resource group, wherein each conversion resource group has a corresponding conversion cost value. The access cost of the user to different resource types is different, and the access cost value corresponding to each conversion resource group can be calculated according to the access cost value and the access times corresponding to each resource type; and taking the sum of the access cost value and the conversion cost value corresponding to each conversion resource group as the final cost value of each conversion resource group, and selecting the conversion resource group with the minimum cost value as the recommended resource. By converting the data resources, the information resources and the knowledge resources of the resource points contained in the difference resources, the dynamic adjustment of the difference resources can be realized, and the conversion resource group with the minimum cost value can be recommended to the user by combining the access cost value of the user to the resources, so that the resource recommendation effect is effectively improved.
Fig. 4 is a schematic hardware structure diagram of an essential computing and inference fusion-oriented cross-modality recommendation apparatus 40 according to an embodiment of the present invention, including:
a memory 41 for storing a computer program;
a processor 42 for executing the computer program to implement the steps of the cross-modality recommendation method oriented to essential computing and reasoning fusion as described in any of the embodiments above.
The embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the cross-modality recommendation method oriented to essential computing and inference fusion described in any of the above embodiments are implemented.
The cross-modal recommendation method and device oriented to essential computing and reasoning fusion and the computer-readable storage medium provided by the embodiment of the invention are described in detail above. The embodiments are described in a progressive manner in the specification, 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 by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the present invention without departing from the principle of the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention.
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 implementation. 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 invention.
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.
Claims (10)
1. A cross-modal recommendation method oriented to essential computing and reasoning fusion is characterized by comprising the following steps:
constructing a data map, an information spectrogram and a knowledge map for the acquired target task by utilizing a pre-stored resource information base;
matching the acquired user resource information with the data map, the information spectrogram and the knowledge map to determine difference resources;
calculating the conversion cost value of the difference resources for resource point conversion according to a pre-established resource fusion cost library; the resource fusion cost library comprises conversion cost values for converting data resources, information resources and knowledge resources; each pair of the difference resources performs the conversion of the resource point once to obtain a conversion resource group, and each conversion resource group has a corresponding conversion cost value;
calculating the access cost value corresponding to each conversion resource group according to the access cost value and the access times corresponding to each resource type;
taking the sum of the access cost value and the conversion cost value corresponding to each conversion resource group as the final cost value of each conversion resource group;
and selecting the conversion resource group with the minimum cost value as a recommended resource.
2. The essential computing and reasoning fusion oriented cross-modal recommendation method according to claim 1, wherein the calculating a conversion cost value for resource point conversion of the difference resource according to a pre-established resource fusion cost library comprises:
setting labels for all resource points contained in the difference resources according to the resource types of the target tasks and the resource types of all resource points in the difference resources; wherein the tags include switchable tags and non-switchable tags;
dividing the resource points provided with the convertible labels into a first resource group; dividing the resource points provided with the non-convertible labels into a second resource group;
performing resource transformation on resource points in the first resource group to obtain a plurality of initial transformation resource groups; wherein, an initial transformation resource group is obtained by executing the transformation of the resource point each time;
sequentially inquiring initial conversion cost values corresponding to all resource points in each initial conversion resource group from the resource fusion cost library to obtain a first conversion cost value corresponding to each initial conversion resource group; judging whether a target conversion resource group with a first conversion cost value smaller than the directly added cost value corresponding to the first resource group exists or not;
when a target conversion resource group with a first conversion cost value smaller than the directly added cost value corresponding to the first resource group exists, combining the target conversion resource group and the second resource group to serve as a conversion resource group, and taking the sum of the first conversion cost value corresponding to the target conversion resource group and the second cost value corresponding to the second resource group as the conversion cost value of the conversion resource group;
and when a target conversion resource group with a first conversion cost value smaller than the directly added cost value corresponding to the first resource group does not exist, combining the first resource group and the second resource group to serve as a conversion resource group, and taking the sum of the directly added cost value corresponding to the first resource group and the second cost value corresponding to the second resource group as the conversion cost value of the conversion resource group.
3. The essential computing and reasoning fusion-oriented cross-modal recommendation method according to claim 2, wherein the sequentially querying, from the resource fusion cost library, initial conversion cost values corresponding to all resource points in each initial conversion resource group to obtain a first conversion cost value corresponding to each initial conversion resource group comprises:
dividing resource points for executing resource conversion in the initial conversion resource group into first type resource points; dividing resource points which do not execute resource conversion in the initial conversion resource group into second type resource points;
inquiring the direct new cost value corresponding to the second type of resource points from the resource fusion cost library;
inquiring the initial conversion cost value corresponding to the first type of resource point from the resource fusion cost library according to the resource type before the first type of resource point performs resource conversion and the resource type after the resource conversion is performed; wherein the initial conversion cost value comprises a temporal cost value, a spatial consumption cost value, a bandwidth consumption cost value, and a composite cost value;
and taking the sum of the directly added cost value and the initial conversion cost value as the first conversion cost value of the initial conversion resource group.
4. The essential computing and reasoning fusion oriented cross-modality recommendation method of claim 1, wherein the matching of the acquired user resource information with the data graph, the information graph and the knowledge graph and the determining of the difference resource comprise:
converting the resource information of the user into a user data map, a user information map and a user knowledge map;
matching the user data map with the data map to obtain data difference resources; matching the user information map with the information map to obtain information difference resources; matching the user knowledge graph with the knowledge graph to obtain knowledge difference resources;
and summarizing the data difference resources, the information difference resources and the knowledge difference resources to be used as difference resources.
5. The essential computing and reasoning fusion-oriented cross-modal recommendation method according to claim 1, wherein the calculating the access cost value corresponding to each conversion resource group according to the access cost value and the access frequency corresponding to each resource type comprises:
determining data access cost values corresponding to all data resources in the conversion resource group according to a preset unit data volume access cost value and the total data amount contained in the conversion resource group;
determining information access cost values corresponding to all information resources in the conversion resource group according to the information access cost values corresponding to all path organization modes;
determining knowledge access cost values corresponding to all knowledge resources in the conversion resource group according to preset single access cost values of various knowledge rules and the types and the number of the knowledge rules contained in the conversion resource group;
and taking the sum of the data access cost value, the information access cost value and the knowledge access cost value as the access cost value of the conversion resource group.
6. The essential computing and reasoning fusion oriented cross-modality recommendation method according to any one of claims 1-4, wherein the constructing of the data graph, the information graph and the knowledge graph for the acquired target task by using the pre-stored resource information base comprises:
decomposing the acquired target task according to a preset classification rule to obtain a plurality of subtasks; each subtask consists of a plurality of track nodes;
obtaining an influence value of the target subtask according to the positive value, the negative value and the required value corresponding to each track node in the target subtask; the target subtask is any one of all the subtasks;
selecting and sequencing the influence values of the subtasks by using a directed graph longest path selection algorithm to obtain an optimal path;
inquiring a pre-stored resource information base to determine resources corresponding to each track node contained in the optimal path; wherein the resources comprise data resources, information resources and knowledge resources;
constructing a data map according to all the data resources; constructing an information map according to all the information resources; and constructing a knowledge graph according to all the knowledge resources.
7. A cross-modal recommendation device oriented to essential computing and reasoning fusion is characterized by comprising a construction unit, a determination unit, a conversion computing unit, an access computing unit, a summary unit and a selection unit;
the construction unit is used for constructing a data map, an information spectrogram and a knowledge map for the acquired target task by utilizing a pre-stored resource information base;
the determining unit is configured to match the acquired user resource information with the data map, the information spectrogram and the knowledge map, and determine a difference resource;
the conversion calculation unit is used for calculating the conversion cost value of the difference resource for resource point conversion according to a pre-established resource fusion cost library; the resource fusion cost library comprises conversion cost values for converting data resources, information resources and knowledge resources; each pair of the difference resources performs the conversion of the resource point once to obtain a conversion resource group, and each conversion resource group has a corresponding conversion cost value;
the access calculation unit is used for calculating the access cost value corresponding to each conversion resource group according to the access cost value and the access frequency corresponding to each resource type;
the summarizing unit is used for taking the sum of the access cost value and the conversion cost value corresponding to each conversion resource group as the final cost value of each conversion resource group;
and the selecting unit is used for selecting the conversion resource group with the minimum cost value as the recommended resource.
8. The essential computing and reasoning fusion oriented cross-modal recommendation device of claim 7, wherein the computing unit comprises a setting subunit, a dividing subunit, a transforming subunit, a querying subunit, a judging subunit, a first as subunit and a second as subunit;
the setting subunit is configured to set a label for each resource point included in the difference resource according to the resource type of the target task and the resource type of each resource point in the difference resource; wherein the tags include switchable tags and non-switchable tags;
the dividing subunit is used for dividing the resource points provided with the convertible labels into a first resource group; dividing the resource points provided with the non-convertible labels into a second resource group;
the transformation sub-unit is used for transforming the resources of the resource points in the first resource group to obtain a plurality of initial transformation resource groups; wherein, an initial transformation resource group is obtained by executing the transformation of the resource point each time;
the query subunit is configured to sequentially query, from the resource fusion cost library, initial conversion cost values corresponding to all resource points in each initial conversion resource group to obtain a first conversion cost value corresponding to each initial conversion resource group;
the judging subunit is configured to judge whether there is a target conversion resource group of which a first conversion cost value is smaller than a directly added cost value corresponding to the first resource group;
the first serving subunit is configured to, when there is a target conversion resource group with a first conversion cost value smaller than a directly added cost value corresponding to the first resource group, combine the target conversion resource group and the second resource group to serve as a conversion resource group, and use a sum of a first conversion cost value corresponding to the target conversion resource group and a second cost value corresponding to the second resource group as a conversion cost value of the conversion resource group;
and the second serving as a subunit, configured to, when there is no target conversion resource group in which the first conversion cost value is smaller than the directly added cost value corresponding to the first resource group, combine the first resource group and the second resource group to serve as a conversion resource group, and use a sum of the directly added cost value corresponding to the first resource group and the second cost value corresponding to the second resource group as the conversion cost value of the conversion resource group.
9. A cross-modal recommendation device oriented to essential computing and reasoning fusion is characterized by comprising:
a memory for storing a computer program;
a processor for executing said computer program for carrying out the steps of the essential computing and reasoning fusion oriented cross-modality recommendation method of any one of claims 1 to 6.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when being executed by a processor, carries out the steps of the inter-modal recommendation method towards intrinsic computing and inference fusion of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010856960.3A CN112000854B (en) | 2020-08-24 | 2020-08-24 | Cross-modal recommendation method and device oriented to essential computing and reasoning fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010856960.3A CN112000854B (en) | 2020-08-24 | 2020-08-24 | Cross-modal recommendation method and device oriented to essential computing and reasoning fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112000854A true CN112000854A (en) | 2020-11-27 |
| CN112000854B CN112000854B (en) | 2021-10-01 |
Family
ID=73470383
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010856960.3A Active CN112000854B (en) | 2020-08-24 | 2020-08-24 | Cross-modal recommendation method and device oriented to essential computing and reasoning fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112000854B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112418428A (en) * | 2020-12-01 | 2021-02-26 | 海南大学 | Cross-modal feature mining method and component based on essential computation |
| CN112782982A (en) * | 2020-12-31 | 2021-05-11 | 海南大学 | Intent-driven essential computation-oriented programmable intelligent control method and system |
| CN112818382A (en) * | 2021-01-13 | 2021-05-18 | 海南大学 | Essential computing-oriented DIKW private resource processing method and component |
| CN112966924A (en) * | 2021-03-02 | 2021-06-15 | 杭州全视软件有限公司 | Data management system and method based on risk map |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104809127A (en) * | 2014-01-26 | 2015-07-29 | 上海联影医疗科技有限公司 | Remote education resource recommendation method and device |
| CN107133371A (en) * | 2017-06-28 | 2017-09-05 | 海南大学 | A kind of spatiotemporal efficiency optimization method towards issued transaction of storage for putting into driving with calculating integrative coordinated adjustment |
| CN107330129A (en) * | 2017-08-16 | 2017-11-07 | 海南大学 | Towards the storage of the input driving of the typing resource issued transaction optimization method integrated with calculating |
| CN107341187A (en) * | 2017-06-07 | 2017-11-10 | 努比亚技术有限公司 | Search processing method, device, equipment and computer-readable storage medium |
-
2020
- 2020-08-24 CN CN202010856960.3A patent/CN112000854B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104809127A (en) * | 2014-01-26 | 2015-07-29 | 上海联影医疗科技有限公司 | Remote education resource recommendation method and device |
| CN107341187A (en) * | 2017-06-07 | 2017-11-10 | 努比亚技术有限公司 | Search processing method, device, equipment and computer-readable storage medium |
| CN107133371A (en) * | 2017-06-28 | 2017-09-05 | 海南大学 | A kind of spatiotemporal efficiency optimization method towards issued transaction of storage for putting into driving with calculating integrative coordinated adjustment |
| CN107330129A (en) * | 2017-08-16 | 2017-11-07 | 海南大学 | Towards the storage of the input driving of the typing resource issued transaction optimization method integrated with calculating |
Non-Patent Citations (1)
| Title |
|---|
| 段玉聪 等: "基于知识图谱的云端个性化测试推荐", 《小型微型计算机系统》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112418428A (en) * | 2020-12-01 | 2021-02-26 | 海南大学 | Cross-modal feature mining method and component based on essential computation |
| CN112418428B (en) * | 2020-12-01 | 2022-04-01 | 海南大学 | Cross-modal feature mining method and component based on essential computation |
| CN112782982A (en) * | 2020-12-31 | 2021-05-11 | 海南大学 | Intent-driven essential computation-oriented programmable intelligent control method and system |
| CN112818382A (en) * | 2021-01-13 | 2021-05-18 | 海南大学 | Essential computing-oriented DIKW private resource processing method and component |
| CN112818382B (en) * | 2021-01-13 | 2022-02-22 | 海南大学 | Essential computing-oriented DIKW private resource processing method and component |
| CN112966924A (en) * | 2021-03-02 | 2021-06-15 | 杭州全视软件有限公司 | Data management system and method based on risk map |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112000854B (en) | 2021-10-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112000854B (en) | Cross-modal recommendation method and device oriented to essential computing and reasoning fusion | |
| JP6414363B2 (en) | Prediction system, method and program | |
| Hasenpusch et al. | Strategic media venturing: Corporate venture capital approaches of TIME incumbents | |
| Md et al. | Dynamic ranking of cloud services for web-based cloud communities: efficient algorithm for rating-based discovery and multi-level ranking of cloud services | |
| Agasisti et al. | Do merger policies increase universities’ efficiency? Evidence from a fuzzy regression discontinuity design | |
| Kardan et al. | A hybrid approach for thread recommendation in MOOC forums | |
| CN111882426A (en) | Business risk classifier training method, device, equipment and storage medium | |
| Liu et al. | A heuristic QoS-aware service selection approach to web service composition | |
| Škare et al. | Measuring economic growth using data envelopment analysis | |
| CN112733035A (en) | Knowledge point recommendation method and device based on knowledge graph, storage medium and electronic device | |
| Ameen et al. | Impact of inspirational motivation on organizational innovation (Administrative innovation, process innovation, and product innovation) | |
| Law et al. | Defining (human) computation | |
| Hüllermeier | Similarity-based inference as evidential reasoning | |
| CN113780365A (en) | Sample generation method and device | |
| Hidayat et al. | Data-driven journalism based on big data analytics: A model development from Indonesia’s experience | |
| CN108550019A (en) | A kind of resume selection method and device | |
| Janoušková et al. | Relevance–a neglected feature of sustainability indicators | |
| CN111062449A (en) | Prediction model training method, interestingness prediction device and storage medium | |
| CN115564578B (en) | Fraud recognition model generation method | |
| CN112463964B (en) | Text classification and model training method, device, equipment and storage medium | |
| Gediga et al. | Approximation quality for sorting rules | |
| Dinga et al. | Sources of macroeconomic instability: implications on foreign direct investment inflow in Sub-Saharan Africa, A PMG/ARDL Approach | |
| Deshmukh et al. | Structural equation modelling of student’s intention towards entrepreneurship in agribusiness | |
| Hernes et al. | Integration of collective knowledge in financial decision support system | |
| Thaha et al. | Cloud Service Provider Selection Using Fuzzy Data Envelopment Analysis Based on SMI Attributes |
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 |