CN114385714B - Multi-sequence fusion method, device and equipment based on feedback and readable storage medium - Google Patents
Multi-sequence fusion method, device and equipment based on feedback and readable storage medium Download PDFInfo
- Publication number
- CN114385714B CN114385714B CN202210039274.6A CN202210039274A CN114385714B CN 114385714 B CN114385714 B CN 114385714B CN 202210039274 A CN202210039274 A CN 202210039274A CN 114385714 B CN114385714 B CN 114385714B
- Authority
- CN
- China
- Prior art keywords
- list
- sorting
- objects
- feedback
- ordered
- 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
- 238000007500 overflow downdraw method Methods 0.000 title claims abstract description 22
- 230000004927 fusion Effects 0.000 claims abstract description 58
- 238000012163 sequencing technique Methods 0.000 claims abstract description 47
- 238000010606 normalization Methods 0.000 claims description 4
- 238000000034 method Methods 0.000 abstract description 15
- 230000008713 feedback mechanism Effects 0.000 abstract description 3
- 238000004891 communication Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2474—Sequence data queries, e.g. querying versioned data
-
- 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
- G06F16/22—Indexing; Data structures therefor; Storage structures
- G06F16/2282—Tablespace storage structures; Management thereof
-
- 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
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Mathematical Physics (AREA)
- Fuzzy Systems (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a multi-row fusion method, a multi-row fusion device and a multi-row fusion device based on feedback, and a readable storage medium. The method comprises the following steps: sorting a plurality of objects to be sorted through at least two sorting models respectively to obtain a sorting list output by each sorting model; fusing all the sequencing lists through equal weights to obtain a fused sequencing list; acquiring feedback information, wherein the feedback information comprises labels of the first K objects in the fusion ordering list; calculating a weight of each ordered list based on the feedback information; and fusing all the sorting lists based on the weight of each sorting list to obtain a new fused sorting list. According to the invention, the weight of each sequencing list is determined based on a feedback mechanism, so that the influence of an inaccurate sequencing list on a final result is reduced, and the accuracy of a new fusion sequencing list finally obtained is improved.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a feedback-based multi-row fusion method, apparatus, device, and readable storage medium.
Background
In a scenario involving ordering of several objects, for example, a scenario ordering the appreciation of a plurality of movies or a scenario ordering the similarity of a plurality of face images to a standard face image, it is often necessary to order several objects using an ordering model.
In the prior art, in order to improve accuracy of the sorting results, a plurality of objects are sorted by a plurality of sorting models, and then a plurality of sorting results are fused in an equal weight manner to obtain a fused sorting result. However, in this manner, if there are poor-performance ranking models among the plurality of ranking models, the accuracy of the obtained fusion ranking result will be low.
Disclosure of Invention
The invention mainly aims to provide a multi-row fusion method, a multi-row fusion device, multi-row fusion equipment and a multi-row fusion readable storage medium based on feedback, and aims to solve the technical problem that in the prior art, a sequencing model with poor performance exists in a plurality of sequencing models, so that the accuracy of an obtained fusion sequencing result is low.
In a first aspect, the present invention provides a feedback-based multi-sequencing fusion method, the feedback-based multi-sequencing fusion method comprising:
Sorting a plurality of objects to be sorted through at least two sorting models respectively to obtain a sorting list output by each sorting model;
fusing all the sequencing lists through equal weights to obtain a fused sequencing list;
Acquiring feedback information, wherein the feedback information comprises labels of the first K objects in the fusion ordering list;
calculating a weight of each ordered list based on the feedback information;
And fusing all the sorting lists based on the weight of each sorting list to obtain a new fused sorting list.
In a second aspect, the present invention also provides a feedback-based multi-sequencing fusion device comprising:
The sorting module is used for sorting a plurality of objects to be sorted through at least two sorting models respectively to obtain a sorting list output by each sorting model;
The first fusion module is used for fusing all the sequencing lists through equal weights to obtain a fused sequencing list;
The acquisition module is used for acquiring feedback information, wherein the feedback information comprises labels of the first K objects in the fusion ordering list;
the calculation module is used for calculating the weight of each sequencing list based on the feedback information;
And the second fusion module is used for fusing all the sorting lists based on the weight of each sorting list to obtain a new fused sorting list.
In a third aspect, the present invention also provides a feedback-based multi-rank fusion device comprising a processor, a memory, and a feedback-based multi-rank fusion program stored on the memory and executable by the processor, wherein the feedback-based multi-rank fusion program, when executed by the processor, implements the steps of the feedback-based multi-rank fusion method as described above.
In a fourth aspect, the present invention further provides a readable storage medium having stored thereon a feedback-based multi-order fusion program, wherein the feedback-based multi-order fusion program, when executed by a processor, implements the steps of the feedback-based multi-order fusion method as described above.
In the invention, a plurality of objects to be sequenced are sequenced through at least two sequencing models respectively, so as to obtain a sequencing list output by each sequencing model; fusing all the sequencing lists through equal weights to obtain a fused sequencing list; acquiring feedback information, wherein the feedback information comprises labels of the first K objects in the fusion ordering list; calculating a weight of each ordered list based on the feedback information; and fusing all the sorting lists based on the weight of each sorting list to obtain a new fused sorting list. According to the invention, the weight of each sequencing list is determined based on a feedback mechanism, so that the influence of an inaccurate sequencing list on a final result is reduced, and the accuracy of a new fusion sequencing list finally obtained is improved.
Drawings
FIG. 1 is a schematic diagram of a hardware architecture of a feedback-based multi-row fusion device according to an embodiment of the present invention;
FIG. 2 is a flow chart of an embodiment of a feedback-based multi-rank fusion method according to the present invention;
fig. 3 is a schematic functional block diagram of an embodiment of a feedback-based multi-row fusion device according to the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
In a first aspect, embodiments of the present invention provide a feedback-based multi-row fusion device, which may be a personal computer (personal computer, PC), a notebook computer, a server, or the like, having a data processing function.
Referring to fig. 1, fig. 1 is a schematic hardware structure of a feedback-based multi-sequencing fusion device according to an embodiment of the present invention. In an embodiment of the present invention, the feedback-based multi-rank fusion device may include a processor 1001 (e.g., central processing unit Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Wherein the communication bus 1002 is used to enable connected communications between these components; the user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard); the network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., wireless FIdelity WIreless-FICAT interface); the memory 1005 may be a high-speed random access memory (random access memory, RAM) or a stable memory (non-volatile memory), such as a disk memory, and the memory 1005 may alternatively be a storage device independent of the processor 1001. Those skilled in the art will appreciate that the hardware configuration shown in fig. 1 is not limiting of the invention and may include more or fewer components than shown, or may combine certain components, or a different arrangement of components.
With continued reference to FIG. 1, an operating system, a network communication module, a user interface module, and a feedback-based multi-row fusion program may be included in memory 1005, which is one type of computer storage medium in FIG. 1. The processor 1001 may invoke the feedback-based multi-row fusion program stored in the memory 1005, and execute the feedback-based multi-row fusion method provided by the embodiment of the present invention.
In a second aspect, embodiments of the present invention provide a feedback-based multi-rank fusion method.
In an embodiment, referring to fig. 2, fig. 2 is a flow chart of an embodiment of a multi-sequencing fusion method based on feedback according to the present invention. As shown in fig. 2, the feedback-based multi-sequencing fusion method includes:
step S10, sorting a plurality of objects to be sorted through at least two sorting models respectively to obtain a sorting list output by each sorting model;
in this embodiment, the objects to be ranked may be movies or face images. It should be noted that this is only a schematic illustration, and does not constitute a limitation on the objects to be sorted.
Taking a plurality of objects to be ranked as a plurality of movies as an example, scoring each movie by a ranking model 1, and ranking the plurality of movies in order from high to low based on a scoring result to obtain a ranking list 1; similarly, each movie is scored by the ranking model 2, and then the movies are ranked in order from high to low based on the scoring result, resulting in the ranked list 2.
Taking a plurality of objects to be ranked as a plurality of face images as an example, scoring the similarity between each face image and a standard face image through a ranking model 1, and then ranking the plurality of face images according to the sequence from high to low based on the scoring result, thereby obtaining a ranking list 1; similarly, similarity between each face image and the standard face image is scored through the ranking model 2, and then the face images are ranked in order from high to low based on the scoring result, so that a ranking list 2 is obtained.
It is readily understood that the number of ranking models is equal to the number of ranking lists.
Step S20, fusing all the sequencing lists through equal weights to obtain a fused sequencing list;
In this embodiment, taking a plurality of objects to be sorted including face images 1 to 10, the sorted list includes a sorted list 1, a sorted list 2 and a sorted list 3 as an example. Referring to table 1, table 1 is a schematic table of ordered list 1:
Table 1 referring to table 2, table 2 is a schematic representation of ordered list 2;
| Face image 1 | 9.5 |
| Face image 7 | 9.1 |
| Face image 2 | 8.8 |
| Face image 9 | 8.7 |
| Face image 8 | 8.2 |
| Face image 5 | 7.7 |
| Face image 4 | 7.3 |
| Face image 6 | 6.8 |
| Face image 3 | 6.5 |
| Face image 10 | 6.3 |
Table 2 referring to table 3, table 3 is a schematic table of ordered list 3:
| Face image 3 | 0.96 |
| Face image 5 | 0.91 |
| Face image 1 | 0.88 |
| Face image 7 | 0.85 |
| Face image 10 | 0.83 |
| Face image 8 | 0.77 |
| Face image 9 | 0.75 |
| Face image 2 | 0.66 |
| Face image 6 | 0.63 |
| Face image 4 | 0.65 |
TABLE 3 Table 3
As shown in tables 1 to 3, since the algorithms adopted by the different ranking models are different, the obtained ranking lists are different, and the scores corresponding to the different ranking lists are different due to the difference of the scoring mechanisms, normalization processing is required to be performed on each ranking list, namely, the score of each face image in each ranking list is changed to a value between 0 and 1. For example, after the normalization processing, the score of the face image 1 in the sorted list 1 becomes 0.98; the score of the face image 1 in the ordered list 2 becomes 0.95; the score of the face image 1 in the ordered list 3 becomes 0.88. Due to the adoption of the equal weight fusion mode, the total score=0.98+0.95+0.88 of the face image 1 can be obtained. And the same is done, so that the total score of the face images 1-10 can be obtained, and the face images 1-10 are ranked again according to the total score of the face images 1-10, and a fusion ranking list is obtained. Referring to table 4, table 4 is a schematic representation of a fused ordered list.
| Face image 1 |
| Face image 5 |
| Face image 3 |
| Face image 7 |
| Face image 8 |
| Face image 6 |
| Face image 10 |
| Face image 9 |
| Face image 2 |
| Face image 4 |
TABLE 4 Table 4
Step S30, obtaining feedback information, wherein the feedback information comprises labels of the first K objects in the fusion ordering list;
In this embodiment, feedback information is obtained, where the feedback information includes labels of the first K objects in the fused ranking list. The value of K is set according to actual needs, for example, according to the number of objects to be sorted.
When the number of the objects to be sorted is not large, the label of each object to be sorted can be predetermined and stored; when the number of the objects to be ordered is large, the first K objects in the fusion ordering list can be output for marking by a user, so that feedback information based on user operation, namely labels of the first K objects in the fusion ordering list, is obtained.
Taking table 4 as an example, the value of K is 5, that is, the tags of the first 5 objects in the fused ordered list are obtained. When the label is a positive label (marked by 1), the object and the standard face image belong to the same face; when the label is a negative label (marked with '0'), it indicates that the object and the standard face image do not belong to the same face.
Step S40, calculating the weight of each ordered list based on the feedback information;
In this embodiment, the feedback information includes labels of the face image 1, the face image 5, the face image 3, the face image 7, and the face image 8, where the labels of the face image 1, the face image 3, and the face image 7 are positive labels, and the labels of the face image 5 and the face image 8 are negative labels.
Further, in an embodiment, step S40 includes:
determining the number of positive labels in the first K objects to be ordered in each ordering list according to the labels of the first K objects; calculating to obtain the score of each sorting list according to the number of positive labels in the front K objects to be sorted in each sorting list; the weight of each sorted list is determined based on the scores of all sorted lists.
In this embodiment, if K is taken 5, the feedback information includes labels of a face image 1, a face image 5, a face image 3, a face image 7 and a face image 8, where the labels of the face image 1, the face image 3 and the face image 7 are positive labels, the labels of the face image 5 and the face image 8 are negative labels, and by combining with tables 1 to 3, it can be determined that the number of labels of the front K (K is taken 5) objects to be ordered in the ordered list 1 is 3, the number of labels of the front 5 objects to be ordered in the ordered list 2 is 2, and the number of labels of the front 5 objects to be ordered in the ordered list 3 is 3.
The scoring rules for the hypothetical ordered list are: the number of positive labels in the first K objects to be sorted in the sorted list, m=the score of the sorted list. Thus, the score of rank list 1 is 3, the score of rank list 2 is 2, and the score of rank list 3 is 3. On the basis, the weight of the ordered list 1 can be determined to be 3/8, the weight of the ordered list 2 is determined to be 2/8, and the weight of the ordered list 3 is determined to be 3/8.
Further, in an embodiment, step S40 includes:
Determining target objects to be ordered, the labels of which are positive labels, according to the labels of the first K objects; obtaining a score set of a target object to be sorted in each sorting list, calculating to obtain a standard deviation of the score set, and taking the reciprocal of the standard deviation as the score of each sorting list; the weight of each sorted list is determined based on the scores of all sorted lists.
In this embodiment, if K is taken 5, the feedback information includes labels of the face image 1, the face image 5, the face image 3, the face image 7 and the face image 8, where the labels of the face image 1, the face image 3 and the face image 7 are positive labels, and the labels of the face image 5 and the face image 8 are negative labels, then the target to-be-sequenced object whose labels are the positive labels includes the face image 1, the face image 3 and the face image 7.
In combination with table 1, the score sets of the face image 1, the face image 3 and the face image 7 in the ordered list 1 are obtained as follows: 98. 96 and 80, calculating the standard deviation of the score set, and then taking the reciprocal of the standard deviation as the score of the ordered list 1; and so on, the score of each ordered list can be obtained. On this basis, the weight of each sorted list may be determined based on the scores of all sorted lists.
And step S50, fusing all the sorting lists based on the weight of each sorting list to obtain a new fused sorting list.
In this embodiment, all the sorted lists are not fused based on equal weights, but fused based on the weights of each sorted list calculated in step S40, to obtain a new fused sorted list.
Further, in an embodiment, step S50 includes:
Normalizing the score of each object to be ordered in each ordered list to obtain a new ordered list; multiplying the score of each object to be ordered in each new ordered list by the weight of each ordered list to obtain the sub-score of each object to be ordered in each new ordered list; adding all the sub-scores corresponding to each object to be sorted to obtain a comprehensive score of each object to be sorted; and sorting all the objects to be sorted based on the comprehensive scores of all the objects to be sorted to obtain a new fused sorting list.
In this embodiment, the score of each object to be sorted in each sorting list is normalized first, and the specific process may refer to the embodiment of step S20, which is not described herein.
For example, after normalization processing, the sub-score of the face image 1 in the ordered list 1 is 0.98; the sub-score of the face image 1 in the ordered list 2 is 0.95; the sub-score of the face image 1 in the ordered list 3 is 0.88; based on the calculation in step S40, it is determined that the weight of the ordered list 1 is 3/8, the weight of the ordered list 2 is 2/8, and the weight of the ordered list 3 is 3/8. Then And the same is done, so that the comprehensive score of each object to be ranked (face images 1-10) can be obtained, and all objects to be ranked are ranked based on the comprehensive scores of all objects to be ranked, so that a new fused ranking list is obtained.
In this embodiment, a plurality of objects to be sorted are sorted by at least two sorting models respectively, so as to obtain a sorting list output by each sorting model; fusing all the sequencing lists through equal weights to obtain a fused sequencing list; acquiring feedback information, wherein the feedback information comprises labels of the first K objects in the fusion ordering list; calculating a weight of each ordered list based on the feedback information; and fusing all the sorting lists based on the weight of each sorting list to obtain a new fused sorting list. According to the embodiment, the weight of each sorting list is determined based on a feedback mechanism, so that the influence of an inaccurate sorting list on a final result is reduced, and the accuracy of a new fusion sorting list finally obtained is improved.
Further, in an embodiment, after step S50, the method further includes:
If new feedback information is received, the new feedback information comprises labels of the first L objects in the fusion ordering list, wherein L is larger than K;
And taking the new feedback information as feedback information, and executing the step of calculating the weight of each sequencing list based on the feedback information.
In this embodiment, after obtaining the new fused ordered list, if new feedback information is received, where the new feedback information includes labels of the first L objects in the fused ordered list, where L is greater than K; then, the new feedback information is used as feedback information, and the steps S40 to S50 are executed in a return mode.
It is easy to understand that if the feedback information includes the labels of the first K objects in the fused ordered list, the obtained new fused ordered list does not conform to the expectations, the feedback information further includes the labels of the first L objects in the fused ordered list, so that the accuracy of the finally obtained fused ordered list is further improved.
In a third aspect, the embodiment of the invention further provides a multi-row fusion device based on feedback.
In an embodiment, referring to fig. 3, fig. 3 is a schematic functional block diagram of an embodiment of a feedback-based multi-sequencing fusion device according to the present invention. As shown in fig. 3, the feedback-based multi-sequencing fusion device includes:
The sorting module 10 is configured to sort a plurality of objects to be sorted through at least two sorting models, so as to obtain a sorting list output by each sorting model;
The first fusing module 20 is configured to fuse all the ordered lists through equal weights to obtain a fused ordered list;
The acquiring module 30 is configured to acquire feedback information, where the feedback information includes tags of the first K objects in the fused ordered list;
a calculation module 40 for calculating a weight of each ordered list based on the feedback information;
And a second fusing module 50, configured to fuse all the sorted lists based on the weight of each sorted list, so as to obtain a new fused sorted list.
Further, in an embodiment, the calculating module 40 is configured to:
determining the number of positive labels in the first K objects to be ordered in each ordering list according to the labels of the first K objects;
Calculating to obtain the score of each sorting list according to the number of positive labels in the front K objects to be sorted in each sorting list;
the weight of each sorted list is determined based on the scores of all sorted lists.
Further, in an embodiment, the calculating module 40 is configured to:
Determining target objects to be ordered, the labels of which are positive labels, according to the labels of the first K objects;
Obtaining a score set of a target object to be sorted in each sorting list, calculating to obtain a standard deviation of the score set, and taking the reciprocal of the standard deviation as the score of each sorting list;
the weight of each sorted list is determined based on the scores of all sorted lists.
Further, in an embodiment, the second fusion module 50 is configured to:
Normalizing the score of each object to be ordered in each ordered list to obtain a new ordered list;
Multiplying the score of each object to be ordered in each new ordered list by the weight of each ordered list to obtain the sub-score of each object to be ordered in each new ordered list;
adding all the sub-scores corresponding to each object to be sorted to obtain a comprehensive score of each object to be sorted;
and sorting all the objects to be sorted based on the comprehensive scores of all the objects to be sorted to obtain a new fused sorting list.
Further, in an embodiment, the feedback-based multi-sequencing fusion device further includes a loop module for:
If new feedback information is received, the new feedback information comprises labels of the first L objects in the fusion ordering list, wherein L is larger than K;
And taking the new feedback information as feedback information, and executing the step of calculating the weight of each sequencing list based on the feedback information.
The function implementation of each module in the feedback-based multi-sequence fusion device corresponds to each step in the feedback-based multi-sequence fusion method embodiment, and the function and implementation process of each module are not described in detail herein.
In a fourth aspect, embodiments of the present invention also provide a readable storage medium.
The readable storage medium of the present invention stores a feedback-based multi-rank fusion program, wherein the feedback-based multi-rank fusion program, when executed by a processor, implements the steps of the feedback-based multi-rank fusion method described above.
The method implemented when the feedback-based multi-sequencing fusion procedure is executed may refer to various embodiments of the feedback-based multi-sequencing fusion method of the present invention, and will not be described herein.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) as described above, comprising several instructions for causing a terminal device to perform the method according to the embodiments of the present invention.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (5)
1. A feedback-based multi-sequencing fusion method, the feedback-based multi-sequencing fusion method comprising:
Sorting a plurality of objects to be sorted through at least two sorting models respectively to obtain a sorting list output by each sorting model;
fusing all the sequencing lists through equal weights to obtain a fused sequencing list;
Acquiring feedback information, wherein the feedback information comprises labels of the first K objects in the fusion ordering list;
calculating a weight of each ordered list based on the feedback information;
fusing all the sorting lists based on the weight of each sorting list to obtain a new fused sorting list;
the step of calculating the weight of each ordered list based on the feedback information includes:
Determining the number of positive labels in the first K objects to be ordered in each ordering list according to the labels of the first K objects; calculating to obtain the score of each sorting list according to the number of positive labels in the front K objects to be sorted in each sorting list; determining the weight of each sorting list according to the scores of all sorting lists;
Or alternatively, the first and second heat exchangers may be,
Determining target objects to be ordered, the labels of which are positive labels, according to the labels of the first K objects; obtaining a score set of a target object to be sorted in each sorting list, calculating to obtain a standard deviation of the score set, and taking the reciprocal of the standard deviation as the score of each sorting list; determining the weight of each sorting list according to the scores of all sorting lists;
the step of fusing all the sorting lists based on the weight of each sorting list to obtain a new fused sorting list comprises the following steps:
Normalizing the score of each object to be ordered in each ordered list to obtain a new ordered list;
Multiplying the score of each object to be ordered in each new ordered list by the weight of each ordered list to obtain the sub-score of each object to be ordered in each new ordered list;
adding all the sub-scores corresponding to each object to be sorted to obtain a comprehensive score of each object to be sorted;
and sorting all the objects to be sorted based on the comprehensive scores of all the objects to be sorted to obtain a new fused sorting list.
2. The feedback-based multi-rank fusion method of claim 1 wherein after the step of fusing all rank lists based on the weight of each rank list to obtain a new fused rank list, further comprising:
If new feedback information is received, the new feedback information comprises labels of the first L objects in the fusion ordering list, wherein L is larger than K;
And taking the new feedback information as feedback information, and executing the step of calculating the weight of each sequencing list based on the feedback information.
3. A feedback-based multi-sequencing fusion device, the feedback-based multi-sequencing fusion device comprising:
The sorting module is used for sorting a plurality of objects to be sorted through at least two sorting models respectively to obtain a sorting list output by each sorting model;
The first fusion module is used for fusing all the sequencing lists through equal weights to obtain a fused sequencing list;
The acquisition module is used for acquiring feedback information, wherein the feedback information comprises labels of the first K objects in the fusion ordering list;
the computing module is used for determining the number of positive labels in the first K objects to be sorted in each sorting list according to the labels of the first K objects; calculating to obtain the score of each sorting list according to the number of positive labels in the front K objects to be sorted in each sorting list; determining the weight of each sorting list according to the scores of all sorting lists;
Or alternatively, the first and second heat exchangers may be,
The target objects to be ordered, of which the labels are positive, are determined according to the labels of the first K objects; obtaining a score set of a target object to be sorted in each sorting list, calculating to obtain a standard deviation of the score set, and taking the reciprocal of the standard deviation as the score of each sorting list; determining the weight of each sorting list according to the scores of all sorting lists;
The second fusion module is used for carrying out normalization processing on the score of each object to be ordered in each ordered list to obtain a new ordered list; multiplying the score of each object to be ordered in each new ordered list by the weight of each ordered list to obtain the sub-score of each object to be ordered in each new ordered list; adding all the sub-scores corresponding to each object to be sorted to obtain a comprehensive score of each object to be sorted; and sorting all the objects to be sorted based on the comprehensive scores of all the objects to be sorted to obtain a new fused sorting list.
4. A feedback-based multi-order fusion device comprising a processor, a memory, and a feedback-based multi-order fusion program stored on the memory and executable by the processor, wherein the feedback-based multi-order fusion program, when executed by the processor, implements the steps of the feedback-based multi-order fusion method of any of claims 1 or 2.
5. A readable storage medium, wherein a feedback-based multi-rank fusion program is stored on the readable storage medium, wherein the feedback-based multi-rank fusion program, when executed by a processor, implements the steps of the feedback-based multi-rank fusion method according to any of claims 1 or 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210039274.6A CN114385714B (en) | 2022-01-13 | 2022-01-13 | Multi-sequence fusion method, device and equipment based on feedback and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210039274.6A CN114385714B (en) | 2022-01-13 | 2022-01-13 | Multi-sequence fusion method, device and equipment based on feedback and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114385714A CN114385714A (en) | 2022-04-22 |
| CN114385714B true CN114385714B (en) | 2024-09-17 |
Family
ID=81202686
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210039274.6A Active CN114385714B (en) | 2022-01-13 | 2022-01-13 | Multi-sequence fusion method, device and equipment based on feedback and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114385714B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115695502B (en) * | 2022-12-15 | 2023-03-10 | 国网浙江省电力有限公司 | Data processing method and device suitable for reliable power communication |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104462550A (en) * | 2014-12-25 | 2015-03-25 | 武汉大学 | Pedestrian re-recognition method based on similarity and dissimilarity fusion ranking optimization |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104200206B (en) * | 2014-09-09 | 2017-04-26 | 武汉大学 | Double-angle sequencing optimization based pedestrian re-identification method |
| US11275749B2 (en) * | 2018-12-31 | 2022-03-15 | International Business Machines Corporation | Enhanced query performance prediction for information retrieval systems |
-
2022
- 2022-01-13 CN CN202210039274.6A patent/CN114385714B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104462550A (en) * | 2014-12-25 | 2015-03-25 | 武汉大学 | Pedestrian re-recognition method based on similarity and dissimilarity fusion ranking optimization |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114385714A (en) | 2022-04-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103870507B (en) | Method and device of searching based on category | |
| CN112966081B (en) | Method, device, equipment and storage medium for processing question and answer information | |
| WO2022048363A1 (en) | Website classification method and apparatus, computer device, and storage medium | |
| CN113326420B (en) | Question retrieval method, device, electronic equipment and medium | |
| CN112988963B (en) | User intention prediction method, device, equipment and medium based on multi-flow nodes | |
| CN112231555A (en) | Recall method, apparatus, device and storage medium based on user portrait label | |
| CN111611374A (en) | Corpus expansion method and device, electronic equipment and storage medium | |
| CN117235238B (en) | Question answering method, question answering device, storage medium and computer equipment | |
| CN113127621B (en) | Method, device, equipment and storage medium for pushing dialogue module | |
| CN107291774B (en) | Error sample identification method and device | |
| CN114175017A (en) | Model construction method, classification method, device, storage medium and electronic equipment | |
| CN114385714B (en) | Multi-sequence fusion method, device and equipment based on feedback and readable storage medium | |
| CN113934848B (en) | Data classification method and device and electronic equipment | |
| US12019671B2 (en) | Locally constrained self-attentive sequential recommendation | |
| CN112529743B (en) | Contract element extraction method, device, electronic equipment and medium | |
| CN110262906B (en) | Interface label recommendation method and device, storage medium and electronic equipment | |
| US11803796B2 (en) | System, method, electronic device, and storage medium for identifying risk event based on social information | |
| CN114661890A (en) | Knowledge recommendation method, device, system and storage medium | |
| CN115248890A (en) | User interest portrait generation method and device, electronic equipment and storage medium | |
| CN110210572B (en) | Image classification method, device, storage medium and equipment | |
| CN112948526A (en) | User portrait generation method and device, electronic equipment and storage medium | |
| CN114676307B (en) | Sequencing model training method, device, equipment and medium based on user retrieval | |
| CN117893862A (en) | Multi-mode article content processing system and control method thereof | |
| CN117195046A (en) | Abnormal text recognition method and related equipment | |
| CN108038735A (en) | Data creation method and device |
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 |