CN113065780A - Task allocation method, device, storage medium and computer equipment - Google Patents
Task allocation method, device, storage medium and computer equipment Download PDFInfo
- Publication number
- CN113065780A CN113065780A CN202110384844.0A CN202110384844A CN113065780A CN 113065780 A CN113065780 A CN 113065780A CN 202110384844 A CN202110384844 A CN 202110384844A CN 113065780 A CN113065780 A CN 113065780A
- Authority
- CN
- China
- Prior art keywords
- user
- task
- users
- target
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
- G06Q10/063114—Status monitoring or status determination for a person or group
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Strategic Management (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Databases & Information Systems (AREA)
- Educational Administration (AREA)
- General Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application relates to the technical field of data analysis and discloses a task allocation method, a device, a storage medium and computer equipment, wherein the method comprises the following steps: acquiring user information of all users and task information of tasks to be processed; respectively extracting characteristic words from the user information of each user and extracting key words from the task information; converting the feature words of each user into a first feature vector and converting the keywords into a second feature vector by using a pre-trained word vector model; respectively calculating cosine distances between the first characteristic vector and the second characteristic vector of each user, and screening out a plurality of target characteristic vectors of which the cosine distances are greater than a preset similarity threshold; and respectively querying the first users corresponding to the target characteristic vectors, screening out second users from the plurality of first users according to a preset screening rule, and distributing the tasks to be processed to the second users. According to the method and the device, the second users matched with the tasks are screened out from a large number of users, and the task distribution accuracy is improved.
Description
Technical Field
The present application relates to the field of data analysis technologies, and in particular, to a task allocation method, apparatus, storage medium, and computer device.
Background
In life, when a user group is huge and tasks need to be allocated to one or more users in the user group for processing, a computer is often used for extracting target users from the user group in a random extraction mode and allocating tasks to be processed to the target users, so as to ensure fairness.
However, when allocating tasks to users for processing, the tasks are generally simply allocated to different users at random, and although this rough allocation method ensures fairness, it is easy to allocate the tasks to inappropriate target users, so that the target users cannot complete the tasks with guaranteed quality, and the allocation accuracy is low.
Disclosure of Invention
The present application provides a task allocation method, a task allocation apparatus, a storage medium, and a computer device, so as to allocate a task to a suitable user and improve task allocation accuracy.
In order to achieve the above object, the present application provides a task allocation method, including:
acquiring user information of all users and task information of tasks to be processed; the user information and the task information are text information;
extracting feature words from the user information of each user respectively, and converting the feature words of each user into first feature vectors by using a pre-trained word vector model; the characteristic words are words representing characteristics of the user in a historical processing task process;
extracting a keyword from the task information, and converting the keyword into a second feature vector by using the word vector model;
respectively calculating the cosine distance between the first characteristic vector and the second characteristic vector of each user, and comparing the cosine distance obtained by calculation with a preset similarity threshold;
screening a plurality of first feature vectors of which the cosine distances are greater than the preset similarity threshold value from the first feature vectors according to a comparison result to obtain a plurality of target feature vectors;
and querying users corresponding to the target feature vectors respectively to obtain a plurality of first users, screening out second users from the plurality of first users according to a preset screening rule, and distributing the tasks to be processed to the second users.
Preferably, the step of screening the second user from the plurality of first users according to a preset screening rule includes:
inquiring user names of the first users, numbering the user names of the first users one by one, giving a unique number to each user name, and generating a user list containing the user names and corresponding numbers; the user names of the user list are arranged according to the sequence of the numbers from small to large;
acquiring the number of users required by the task to be processed, randomly selecting the number of the users from the user list, acquiring the number at the tail from the current rest numbers of the user list after selecting one number each time, filling the number in a vacancy corresponding to the selected number, and then performing next number selection;
and taking the number selected each time as a target number, inquiring a user name corresponding to the target number from the user list to obtain a target user name, and taking a first user corresponding to the target user name as the second user.
Preferably, the step of randomly selecting the numbers in the number of the users from the user list, and after selecting one number each time, acquiring a last number from the currently remaining numbers in the user list, filling in a vacancy corresponding to the selected number, and then performing next number selection includes:
judging whether the number of the users is more than 1;
if yes, randomly selecting a first number from the user list when the number is selected for the first time, acquiring a number ranked at the tail from the currently remaining numbers in the user list, and filling the number in a vacancy corresponding to the first number;
and when the number is selected for the second time, randomly selecting a second number from the user list with the filled vacancy, acquiring a number at the tail from the currently remaining numbers in the user list, filling the vacancy corresponding to the second number, and repeating the steps until the number of randomly selected times is the same as the number of the users.
Further, after the step of assigning a unique number to each user name, the method further includes:
generating an array according to the number of each user name, establishing an array index of each number and the corresponding user name for the array, and storing the array index in the user list;
the step of obtaining the target user name by querying the user name corresponding to the target number from the user list includes:
and inquiring the user name corresponding to each target number by using the array index of the user list, and taking the user name corresponding to each target number as the target user name.
Further, the task to be processed is a monitoring task for monitoring a unit, and before the step of obtaining the number of users required by the task to be processed, the method further includes:
randomly extracting a unit to be monitored from a plurality of units according to a random rule;
and formulating a monitoring task for the unit to be monitored to obtain the task to be processed, and determining the number of users required by the task to be processed according to the number of the unit to be monitored.
Preferably, the step of screening the second user from the plurality of first users according to a preset screening rule includes:
acquiring all historical processing tasks of a first user, and screening out the historical processing tasks which accord with the task type of the task to be processed from all the historical processing tasks to obtain a target historical processing task;
inquiring the score value of the first user when the first user processes the target historical processing task from a pre-constructed score list; the scoring list stores scoring values which are evaluated by a system after each user processes a task;
and taking the first user with the score value higher than a preset score value as a second user.
Further, the task to be processed includes an inspection task, and after the step of allocating the task to be processed to the second user, the method further includes:
receiving the uploaded inspection data filled by the second user according to a preset standard format; the inspection data are processing data of the second user aiming at the task to be processed, and the inspection data comprise inspection items, inspection contents, deduction conditions of the inspection contents and existing problems;
and generating an inspection result according to the inspection data.
The present application further provides a task allocation apparatus, including:
the acquisition module is used for acquiring user information of all users and task information of the tasks to be processed; the user information and the task information are text information;
the first conversion module is used for extracting feature words from the user information of each user respectively and converting the feature words of each user into first feature vectors by using a pre-trained word vector model; the characteristic words are words representing characteristics of the user in a historical processing task process;
the second conversion module is used for extracting keywords from the task information and converting the keywords into second feature vectors by using the word vector model;
the calculating module is used for calculating the cosine distance between the first characteristic vector and the second characteristic vector of each user respectively and comparing the cosine distance obtained by calculation with a preset similarity threshold;
the screening module is used for screening a plurality of first feature vectors of which the cosine distances are greater than the preset similarity threshold value from the first feature vectors according to the comparison result to obtain a plurality of target feature vectors;
and the distribution module is used for respectively inquiring the users corresponding to the target characteristic vectors to obtain a plurality of first users, screening out second users from the first users according to a preset screening rule, and distributing the tasks to be processed to the second users.
The present application further provides a computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the steps of any of the above methods when executing the computer program.
The present application also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, carries out the steps of the method of any of the above.
According to the task allocation method, the device, the storage medium and the computer equipment, user information of all users and task information of tasks to be processed are obtained firstly; respectively extracting characteristic words from the user information of each user and extracting key words from the task information; converting the feature words of each user into a first feature vector and converting the keywords into a second feature vector by using a pre-trained word vector model; then, respectively calculating cosine distances between the first characteristic vector and the second characteristic vector of each user, and screening a plurality of first characteristic vectors of which the cosine distances are greater than a preset similarity threshold value from the first characteristic vectors to obtain a plurality of target characteristic vectors; the method comprises the steps of respectively inquiring users corresponding to a plurality of target characteristic vectors to obtain a plurality of first users, screening out second users from the plurality of first users according to a preset screening rule, and distributing tasks to be processed to the second users, so that the second users matched with the tasks are screened out from a large number of users by vector matching of characteristic words of each user and key words of task information in a vector mode, the tasks are distributed to the second users, accurate distribution is achieved, and task distribution accuracy is improved.
Drawings
Fig. 1 is a schematic flowchart of a task allocation method according to an embodiment of the present application;
FIG. 2 is a block diagram illustrating a task allocation apparatus according to an embodiment of the present disclosure;
fig. 3 is a block diagram illustrating a structure of a computer device according to an embodiment of the present application.
The implementation, functional features and advantages of the objectives of the present application will be further explained with reference to the accompanying drawings.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The present application provides a task allocation method, which is used to solve the problem that when a task is randomly allocated at present, the task is allocated to an inappropriate target user, so that the target user cannot complete the task with guaranteed quality and low allocation accuracy, in an embodiment, as shown in fig. 1, the task allocation method includes:
s1, acquiring user information of all users and task information of the tasks to be processed; the user information and the task information are text information;
the user information and the task information are text information represented in a text form, and when the user information or the task information is picture information, video information or voice information, the picture information, the video information or the voice information can be converted into the text information form in advance. The user information may include a user name, a number of historical processing tasks, a type of skilled processing task, and the like. The task information may include task content, difficulty factors, domain of belongings, task type, and so on.
S2, respectively extracting feature words from the user information of each user, and converting the feature words of each user into a first feature vector by using a pre-trained word vector model; the characteristic words are words representing characteristics of the user in a historical processing task process;
the word segmentation algorithm can be used for carrying out word segmentation on the user information, and then the characteristic words of the user information are determined, wherein the characteristic words are phrases with substantial meanings in the user information, such as words with parts of speech, such as adjectives, verbs or nouns, and represent characteristics of the user in the history processing task process, such as the number of history processing tasks, words with good conditions of the history processing tasks, and expression words with good processing tasks. The word vector model is a model considering the position relationship of words. Through training of a large amount of corpora, each word is mapped into a high-dimensional (more than thousands of dimensions and tens of thousands of dimensions) vector, and the relation between two words can be judged in a cosine solving mode, for example, in a word vector model of Jane and Bob in an example sentence, the cosine values of Jane and Bob are probably close to 1 because the two words are both names of people, and the cosine values of Shenzhen and Bob are probably close to 0 because one word is a name of a person and one word is a name of a place. Word2vec is commonly used to form a word vector model, and a neural network model based on CBOW and Skip-Gram algorithms is adopted at the bottom layer of the word vector model.
In an embodiment, when a plurality of feature words are extracted from user information of any one user, the plurality of feature words are respectively converted into feature vectors by using a pre-trained word vector model to obtain a plurality of feature vectors of the user, and an average vector of the plurality of feature vectors is used as the first feature vector.
S3, extracting keywords from the task information, and converting the keywords into second feature vectors by using the word vector model;
in the step, the word segmentation algorithm can be utilized to segment the task information, and then the keyword of the task information is determined, wherein the keyword is a phrase with substantial meaning in the task information, such as an adjective word and a noun, and the keyword represents the characteristics of the task to be processed, such as a task type, a task difficulty coefficient and a word in the field to which the task belongs. And then, converting the keywords into a vector form by using the word vector model to which the keywords belong to so as to obtain a second feature vector. The word vector model used for converting the first feature vector and the second feature vector may be the same word vector model or different word vector models.
In an embodiment, when a plurality of keywords are extracted from the task information, the plurality of keywords are respectively converted into feature vectors by using a pre-trained word vector model to obtain a plurality of feature vectors of the task information, and an average vector of the plurality of feature vectors is used as the second feature vector.
S4, respectively calculating cosine distances of the first feature vectors and the second feature vectors of each user, and comparing the cosine distances obtained through calculation with a preset similarity threshold;
in the step, the cosine distance between the first characteristic vector and the second characteristic vector of each user is calculated, and the cosine distance is compared with a preset similarity threshold value to evaluate whether the user is matched with the task to be processed, wherein the preset similarity threshold value is 0.8-1.
S5, screening a plurality of first feature vectors of which the cosine distances are larger than the preset similarity threshold value from the first feature vectors according to the comparison result to obtain a plurality of target feature vectors;
in the step, a plurality of first feature vectors with cosine distances larger than a preset similarity threshold are screened out from the first feature vectors and are used as a plurality of target feature vectors, namely, the cosine distances of the plurality of target feature vectors are larger than the preset similarity threshold, and the plurality of target feature vectors are vectors corresponding to feature words of the user which are matched with the task to be processed in a comparison mode.
S6, users corresponding to the target feature vectors are inquired respectively to obtain a plurality of first users, a second user is screened from the first users according to a preset screening rule, and the task to be processed is distributed to the second user.
After the target characteristic vector is obtained, a user corresponding to the target characteristic vector is inquired and used as a first user, the first user is a user matched with the task to be processed, so that the first user which is relatively consistent with the task to be processed is screened out from all users, then the first user is randomly extracted from the first user according to a preset screening rule, a second user is obtained, the task to be processed is distributed to the second user, and fairness and accuracy of task distribution are guaranteed. For example, when the task to be processed is monitoring of food such as rice, law enforcement officers who are mainly responsible for monitoring the safety of mechanical equipment do not conform to the task type of the current task to be processed, and the law enforcement officers who are responsible for the task need to be rejected and screened out.
Taking personnel allocation of supervision tasks as an example, the supervision tasks are used as an implementation means of intelligent food safety supervision and inspection, and dual-random is an important supervision mode, mainly refers to that supervision objects and law enforcement personnel of the tasks randomly generate. When the existing double random mode is used for extracting the supervision object and the law enforcement officers, the consistency of affairs in the supervision object and the law enforcement officers cannot be ensured, and the extracted law enforcement officers can generate the phenomenon of repeated extraction. For example, assuming there are ten thousand law enforcement officers, law enforcement officers are now randomly drawn from the ten thousand officers and each randomly drawn officer is not repeatable. The traditional implementation mode is to generate a random number r of 1 to 10000, then take out the corresponding law enforcement officer according to r, judge whether the law enforcement officer repeats (random number collision) by comparing whether the random number generated each time is extracted, if the generated r has the random number collision, r needs to be regenerated, and the collision probability of the random number increases along with the increase of the generated random number, the probability of extracting non-heavy r becomes smaller and smaller (10000/10000- >1/10000), and along with the increase of the extracted data volume, the efficiency influence degree will be more obvious, and the generation efficiency of the double random supervision task is influenced.
Therefore, when the first user is screened, after one number is selected each time, the method obtains the largest number from the numbers corresponding to the remaining first users in the user list, fills the vacancy corresponding to the selected number, and then extracts the number to avoid the influence on the distribution efficiency caused by the idle work of repeatedly extracting the same user.
Specifically, in an embodiment, in this step S6, the step of screening the second user from the plurality of first users according to a preset screening rule may specifically include:
s61, inquiring user names of the first users, numbering the user names of the first users one by one, giving a unique number to each user name, and generating a user list containing the user names and corresponding numbers; the user names of the user list are arranged according to the sequence of the numbers from small to large;
the method comprises the steps of obtaining user names of a plurality of first users obtained through preliminary screening, numbering all the user names of the first users respectively, giving a unique number to each user name, enabling the number to correspond to the user names one by one, numbering the user names one by one according to the sequence of Arabic numbers from small to large during specific numbering, generating a user list containing the user names and the corresponding numbers, and arranging the user names according to the sequence of the numbers from small to large. For example, if the first user shares four law enforcement officers, the first user may sequentially have numbers of a, b, c, d, i.e., the number of a is 1, the number of b is 2, the number of c is 3, and the number of d is 4, and the numbers are arranged from top to bottom or from left to right in the user list, so that the larger number is arranged behind the larger number.
S62, acquiring the number of users needed by the task to be processed, randomly selecting the number of the users from the user list, acquiring the number ranked at the tail from the current rest numbers of the user list after selecting one number each time, filling the vacancy corresponding to the selected number, and then performing next number selection;
the number of the tasks to be processed is obtained, the required number of the users is determined according to the number of the tasks to be processed, and the number of the users is randomly selected from a user list to obtain a target number. Specifically, during each extraction, a number is randomly extracted from the user list, and after each extraction, a vacancy is left in the selected number, so that before the next extraction, the vacancy corresponding to the selected number needs to be filled in from the number arranged at the end currently, then a number is continuously extracted from the user name list with the number filled, and the operation is repeated until the number of the user is extracted.
In an embodiment, the to-be-processed task is a monitoring task for monitoring a unit, and before the step of obtaining the number of users required by the to-be-processed task in step S62, the method may further include:
d60, randomly extracting units to be monitored from the multiple units according to a random rule;
and D61, establishing a monitoring task for the unit to be monitored to obtain the task to be processed, and determining the number of users required by the task to be processed according to the number of the unit to be monitored.
In the embodiment, a 'double random and one public' mode is adopted, namely, the inspection object is randomly extracted in the supervision process, the law enforcement and inspection personnel are randomly selected, and the extraction condition and the inspection result are timely disclosed to the society. Specifically, when the unit to be monitored is extracted, the unit to be monitored is extracted randomly from the multiple units according to a random rule, the monitoring task is formulated for each unit to be monitored, the task to be processed is obtained, the number of users required by the task to be processed is determined according to the number of the unit to be monitored, the unit to be monitored is extracted randomly, and the monitoring effect is improved. For example, when the number of units to be monitored is 3, the number of users required for the tasks to be processed is 9 according to the ratio of 1: 3.
As a preferred embodiment, in this step S62, the randomly selecting the numbers in the user number from the user list, and after selecting one number each time, acquiring the last-ranked number from the currently remaining numbers in the user list, filling the vacancy corresponding to the selected number, and then performing the next number selection step may specifically include:
s621, judging whether the number of the users is greater than 1;
when the number of the users is 1, only one number is randomly extracted from the user list to serve as a target number, and the user corresponding to the target number serves as a person for executing the task to be processed.
S622, if yes, randomly selecting a first number from the user list when the number is selected for the first time, acquiring a number arranged at the tail from the currently remaining numbers in the user list, and filling the number in a vacancy corresponding to the first number;
when the number of users is greater than 1, first extracting is performed, a number is randomly selected from a user list to serve as a target number for first number selection, and after the first number is selected, a vacancy is left in the selected number, a number ranked at the end is obtained from the currently remaining numbers in the user list, and the number ranked at the end and a corresponding user name are filled in the vacancy corresponding to the first number, which is further described in the above embodiment, assuming that after the number 2 of the second is selected, the number ranked at the end is the number 4 of the dice, the number 4 of the dice and the dice are filled in the vacancy of the number 2, and next selection is performed.
And S623, when selecting the number for the second time, randomly selecting a second number from the user list with the filled vacancy, acquiring a number at the tail from the currently remaining numbers in the user list, filling the vacancy corresponding to the second number, and repeating the steps until the number of randomly selected times is the same as the number of the users.
And after filling the vacant positions corresponding to the selected numbers, carrying out second number selection from the filled user list, randomly selecting the second number as a target number selected by the second number, acquiring the numbers ranked at the tail from the currently remaining numbers in the user list, filling the numbers ranked at the tail on the vacant positions corresponding to the second number, carrying out next selection, and repeating the steps until the target numbers with the same number as the number of the users are extracted. For example, in the second number selection, if the number 1 of the first is selected, the number 3 of the number c listed at the end of the current user list is the number 3 of the number c, the number 3 of the number c and the number c are filled in the empty space of the number 1, and the next selection is performed again to avoid repeatedly extracting the same number.
And S63, taking the number selected each time as a target number, inquiring a user name corresponding to the target number from the user list to obtain a target user name, and taking the first user corresponding to the target user name as the second user.
In the step, all the extracted numbers are used as target numbers, the user name corresponding to each number is inquired from a user list according to the target numbers to obtain the target user name, the first user corresponding to the target user name is used as a second user, the task to be processed is distributed to the second user, and random monitoring work is completed.
Therefore, the number collision can be avoided, the second user corresponding to the number extracted every time is ensured to be unrepeated, repeated extraction is avoided, and the generation efficiency of the double random supervision task is improved. In addition, the probability of each second user is changed according to the number of times of extraction, but the probability of each second user in each extraction is kept unchanged, and all second users can be extracted by generating numbers n times at most, so that the random extraction efficiency is remarkably improved.
In an embodiment, in step S61, after the step of assigning a unique number to each user name, the method may further include:
generating an array according to the number of each user name, establishing an array index of each number and the corresponding user name for the array, and storing the array index in the user list;
at this time, in step S63, the step of querying the user name corresponding to the target number from the user list to obtain the target user name may specifically include:
s631, querying the user name corresponding to each target number by using the array index of the user list, and taking the user name corresponding to each target number as the target user name.
In this embodiment, an array is generated according to the number of each user name, an array index of each number in the array and the corresponding user name is established, and the array index is stored in the user list, where the array index may be used to query the user name corresponding to each number, and therefore when a target number is extracted, the user name corresponding to each target number may be queried from the array index of the user list, and the user name corresponding to each target number is used as the target user name. Specifically, taking the user as a supervisor as an example, all supervisors participating in random drawing can be put into an array, and if n supervisors exist, the expression form of the supervisor array(s) is as follows: s ═ S1, S2, S3, S4, …, Sn ], the corresponding array index is [0,1,2,3, …, n-1], the number of times of generating a random number is represented by i, initially i ═ 0, each time a random integer in the [0, n-1-i ] interval is generated, and at the same time i +1, if the generated random number is 3, the supervisor S4 with the index of 3 in S is fetched, and S4 is the supervisor of this random fetch. Placing s [ n-1-i ] at s [ r ], setting s [ n-1-i ] to null, i.e., s [ r ] ═ s [ n-1-i ], s [ n-1-i ] ═ null, when i is 0: and judging whether the supervisor needs to be selected randomly or not, and only when i < ═ n, continuing the extraction until the target numbers with the same number as that of the users are extracted, and finishing the extraction. Where n represents the number of users.
In another embodiment, in this step S6, the step of screening the second user from the plurality of first users according to a preset screening rule includes:
c61, acquiring all historical processing tasks of the first user, and screening out the historical processing tasks which accord with the task type of the to-be-processed task from all the historical processing tasks to obtain a target historical processing task;
c62, inquiring the score value of the first user when processing the target historical processing task from a pre-constructed score list; the scoring list stores scoring values which are evaluated by a system after each user processes a task;
and C63, taking the first user with the score value higher than the preset score value as a second user.
In this embodiment, after the user has processed the tasks in the past, the system may score the tasks processed by the user, and screen out target historical processing tasks that meet the task type from all the historical processing tasks, accumulate the scores of the target historical processing tasks and obtain the score value after averaging, then sort the user names according to the score value from high to low, and screen out the top N users or the users with the score value higher than the preset value from the first user as the second user according to the sorting result, so as to screen out suitable monitoring personnel. Wherein N is a positive integer greater than or equal to 1.
In an embodiment, in step S6, the step of allocating the to-be-processed task to the second user may further include:
receiving the uploaded inspection data filled by the second user according to a preset standard format; the inspection data are processing data of the second user aiming at the task to be processed, and the inspection data comprise inspection items, inspection contents, deduction conditions of the inspection contents and existing problems;
and generating an inspection result according to the inspection data.
In this embodiment, after the second user is determined, a task to be processed is allocated to each second user, so that the second user completes the task to be processed, and after the task is completed by the second user, the inspection data is filled according to a standard format, and after the inspection data is inspected and corrected, an inspection report is generated by using the corrected inspection data. The inspection report may contain information of the type of the monitoring task, the name of the monitoring unit, whether the monitoring is up to standard, and the like.
To sum up, the task allocation method of the present application first obtains user information of all users and task information of tasks to be processed; respectively extracting characteristic words from the user information of each user and extracting key words from the task information; converting the feature words of each user into a first feature vector and converting the keywords into a second feature vector by using a pre-trained word vector model; then, respectively calculating cosine distances between the first characteristic vector and the second characteristic vector of each user, and screening a plurality of first characteristic vectors of which the cosine distances are greater than a preset similarity threshold value from the first characteristic vectors to obtain a plurality of target characteristic vectors; the method comprises the steps of respectively inquiring users corresponding to a plurality of target characteristic vectors to obtain a plurality of first users, screening out second users from the plurality of first users according to a preset screening rule, and distributing tasks to be processed to the second users, so that the second users matched with the tasks are screened out from a large number of users by vector matching of characteristic words of each user and key words of task information in a vector mode, the tasks are distributed to the second users, accurate distribution is achieved, and task distribution accuracy is improved.
Referring to fig. 2, an embodiment of the present application further provides a task allocation apparatus, including:
the acquisition module 1 is used for acquiring user information of all users and task information of tasks to be processed; the user information and the task information are text information;
the user information and the task information are text information represented in a text form, and when the user information or the task information is picture information, video information or voice information, the picture information, the video information or the voice information can be converted into the text information form in advance. The user information may include a user name, a number of historical processing tasks, a type of skilled processing task, and the like. The task information may include task content, difficulty factors, domain of belongings, task type, and so on.
The first conversion module 2 is used for extracting feature words from the user information of each user respectively and converting the feature words of each user into first feature vectors by using a pre-trained word vector model; the characteristic words are words representing characteristics of the user in a historical processing task process;
the module can perform word segmentation on the user information by using a word segmentation algorithm, and then determine a feature word of the user information, wherein the feature word is a word group with substantial meaning in the user information, such as words with parts of speech, such as adjectives, verbs or nouns, and the feature word represents features of a user in a history processing task process, such as the number of history processing tasks, words with good conditions of the history processing tasks, and expression words with good intentions in processing task types. The word vector model is a model considering the position relationship of words. Through training of a large amount of corpora, each word is mapped into a high-dimensional (more than thousands of dimensions and tens of thousands of dimensions) vector, and the relation between two words can be judged in a cosine solving mode, for example, in a word vector model of Jane and Bob in an example sentence, the cosine values of Jane and Bob are probably close to 1 because the two words are both names of people, and the cosine values of Shenzhen and Bob are probably close to 0 because one word is a name of a person and one word is a name of a place. Word2vec is commonly used to form a word vector model, and a neural network model based on CBOW and Skip-Gram algorithms is adopted at the bottom layer of the word vector model.
In an embodiment, when a plurality of feature words are extracted from user information of any one user, the plurality of feature words are respectively converted into feature vectors by using a pre-trained word vector model to obtain a plurality of feature vectors of the user, and an average vector of the plurality of feature vectors is used as the first feature vector.
The second conversion module 3 is used for extracting keywords from the task information and converting the keywords into second feature vectors by using the word vector model;
the module can perform word segmentation on the task information by using a word segmentation algorithm, and then determines a keyword of the task information, wherein the keyword is a phrase with substantial meaning in the task information, such as an adjective word and a noun, and the keyword represents characteristics of a task to be processed, such as a task type, a task difficulty coefficient and a word in a field to which the task belongs. And then, converting the keywords into a vector form by using the word vector model to which the keywords belong to so as to obtain a second feature vector. The word vector model used for converting the first feature vector and the second feature vector may be the same word vector model or different word vector models.
In an embodiment, when a plurality of keywords are extracted from the task information, the plurality of keywords are respectively converted into feature vectors by using a pre-trained word vector model to obtain a plurality of feature vectors of the task information, and an average vector of the plurality of feature vectors is used as the second feature vector.
The calculating module 4 is configured to calculate a cosine distance between the first feature vector and the second feature vector of each user, and compare the cosine distance obtained through calculation with a preset similarity threshold;
the module calculates the cosine distance between the first eigenvector and the second eigenvector of each user, and compares the cosine distance with a preset similarity threshold value to evaluate whether the user is matched with the task to be processed, wherein the preset similarity threshold value is 0.8-1.
The screening module 5 is configured to screen a plurality of first feature vectors of which cosine distances are greater than the preset similarity threshold from the first feature vectors according to the comparison result, so as to obtain a plurality of target feature vectors;
in the step, a plurality of first feature vectors with cosine distances larger than a preset similarity threshold are screened out from the first feature vectors and are used as a plurality of target feature vectors, namely, the cosine distances of the plurality of target feature vectors are larger than the preset similarity threshold, and the plurality of target feature vectors are vectors corresponding to feature words of the user which are matched with the task to be processed in a comparison mode.
And the distribution module 6 is configured to query users corresponding to the target feature vectors respectively to obtain a plurality of first users, screen out a second user from the plurality of first users according to a preset screening rule, and distribute the task to be processed to the second user.
After the target characteristic vector is obtained, a user corresponding to the target characteristic vector is inquired and used as a first user, the first user is a user matched with the task to be processed, so that the first user which is relatively consistent with the task to be processed is screened out from all users, then the first user is randomly extracted from the first user according to a preset screening rule, a second user is obtained, the task to be processed is distributed to the second user, and fairness and accuracy of task distribution are guaranteed. For example, when the task to be processed is monitoring of food such as rice, law enforcement officers who are mainly responsible for monitoring the safety of mechanical equipment do not conform to the task type of the current task to be processed, and the law enforcement officers who are responsible for the task need to be rejected and screened out.
As described above, it can be understood that each component of the task allocation apparatus provided in this application may implement the function of any one of the task allocation methods described above, and the detailed structure is not described again.
Referring to fig. 3, a computer device, which may be a server and whose internal structure may be as shown in fig. 3, is also provided in the embodiment of the present application. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the computer designed processor is used to provide computational and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of task allocation.
The processor executes the task allocation method, and the method comprises the following steps:
acquiring user information of all users and task information of tasks to be processed; the user information and the task information are text information;
extracting feature words from the user information of each user respectively, and converting the feature words of each user into first feature vectors by using a pre-trained word vector model; the characteristic words are words representing characteristics of the user in a historical processing task process;
extracting a keyword from the task information, and converting the keyword into a second feature vector by using the word vector model;
respectively calculating the cosine distance between the first characteristic vector and the second characteristic vector of each user, and comparing the cosine distance obtained by calculation with a preset similarity threshold;
screening a plurality of first feature vectors of which the cosine distances are greater than the preset similarity threshold value from the first feature vectors according to a comparison result to obtain a plurality of target feature vectors;
and querying users corresponding to the target feature vectors respectively to obtain a plurality of first users, screening out second users from the plurality of first users according to a preset screening rule, and distributing the tasks to be processed to the second users.
An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, the computer program, when executed by a processor, implementing a task allocation method, including the steps of:
acquiring user information of all users and task information of tasks to be processed; the user information and the task information are text information;
extracting feature words from the user information of each user respectively, and converting the feature words of each user into first feature vectors by using a pre-trained word vector model; the characteristic words are words representing characteristics of the user in a historical processing task process;
extracting a keyword from the task information, and converting the keyword into a second feature vector by using the word vector model;
respectively calculating the cosine distance between the first characteristic vector and the second characteristic vector of each user, and comparing the cosine distance obtained by calculation with a preset similarity threshold;
screening a plurality of first feature vectors of which the cosine distances are greater than the preset similarity threshold value from the first feature vectors according to a comparison result to obtain a plurality of target feature vectors;
and querying users corresponding to the target feature vectors respectively to obtain a plurality of first users, screening out second users from the plurality of first users according to a preset screening rule, and distributing the tasks to be processed to the second users.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium provided herein and used in the examples may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), double-rate SDRAM (SSRSDRAM), Enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and bus dynamic RAM (RDRAM).
To sum up, the most beneficial effect of this application lies in:
according to the task allocation method, the device, the storage medium and the computer equipment, user information of all users and task information of tasks to be processed are obtained firstly; respectively extracting characteristic words from the user information of each user and extracting key words from the task information; converting the feature words of each user into a first feature vector and converting the keywords into a second feature vector by using a pre-trained word vector model; then, respectively calculating cosine distances between the first characteristic vector and the second characteristic vector of each user, and screening a plurality of first characteristic vectors of which the cosine distances are greater than a preset similarity threshold value from the first characteristic vectors to obtain a plurality of target characteristic vectors; the method comprises the steps of respectively inquiring users corresponding to a plurality of target characteristic vectors to obtain a plurality of first users, screening out second users from the plurality of first users according to a preset screening rule, and distributing tasks to be processed to the second users, so that the second users matched with the tasks are screened out from a large number of users by vector matching of characteristic words of each user and key words of task information in a vector mode, the tasks are distributed to the second users, accurate distribution is achieved, and task distribution accuracy is improved.
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, apparatus, article, or method 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, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that includes the element.
The above description is only a preferred embodiment of the present application, and not intended to limit the scope of the present application, and all modifications of equivalent structures and equivalent processes, which are made by the contents of the specification and the drawings of the present application, or which are directly or indirectly applied to other related technical fields, are also included in the scope of the present application.
Claims (10)
1. A task allocation method, comprising:
acquiring user information of all users and task information of tasks to be processed; the user information and the task information are text information;
extracting feature words from the user information of each user respectively, and converting the feature words of each user into first feature vectors by using a pre-trained word vector model; the characteristic words are words representing characteristics of the user in a historical processing task process;
extracting a keyword from the task information, and converting the keyword into a second feature vector by using the word vector model;
respectively calculating the cosine distance between the first characteristic vector and the second characteristic vector of each user, and comparing the cosine distance obtained by calculation with a preset similarity threshold;
screening a plurality of first feature vectors of which the cosine distances are greater than the preset similarity threshold value from the first feature vectors according to a comparison result to obtain a plurality of target feature vectors;
and querying users corresponding to the target feature vectors respectively to obtain a plurality of first users, screening out second users from the plurality of first users according to a preset screening rule, and distributing the tasks to be processed to the second users.
2. The method according to claim 1, wherein the step of filtering out the second user from the plurality of first users according to a preset filtering rule comprises:
inquiring user names of the first users, numbering the user names of the first users one by one, giving a unique number to each user name, and generating a user list containing the user names and corresponding numbers; the user names of the user list are arranged according to the sequence of the numbers from small to large;
acquiring the number of users required by the task to be processed, randomly selecting the number of the users from the user list, acquiring the number at the tail from the current rest numbers of the user list after selecting one number each time, filling the number in a vacancy corresponding to the selected number, and then performing next number selection;
and taking the number selected each time as a target number, inquiring a user name corresponding to the target number from the user list to obtain a target user name, and taking a first user corresponding to the target user name as the second user.
3. The method of claim 2, wherein the step of randomly selecting the number of the users from the user list, acquiring the last number from the currently remaining numbers in the user list after selecting one number each time, filling the vacancy corresponding to the selected number, and performing the next number selection comprises:
judging whether the number of the users is more than 1;
if yes, randomly selecting a first number from the user list when the number is selected for the first time, acquiring a number ranked at the tail from the currently remaining numbers in the user list, and filling the number in a vacancy corresponding to the first number;
and when the number is selected for the second time, randomly selecting a second number from the user list with the filled vacancy, acquiring a number at the tail from the currently remaining numbers in the user list, filling the vacancy corresponding to the second number, and repeating the steps until the number of randomly selected times is the same as the number of the users.
4. The method of claim 2, wherein the step of assigning a unique number to each username is followed by the step of:
generating an array according to the number of each user name, establishing an array index of each number and the corresponding user name for the array, and storing the array index in the user list;
the step of obtaining the target user name by querying the user name corresponding to the target number from the user list includes:
and inquiring the user name corresponding to each target number by using the array index of the user list, and taking the user name corresponding to each target number as the target user name.
5. The method according to claim 2, wherein the task to be processed is a monitoring task for monitoring a unit, and before the step of obtaining the number of users required by the task to be processed, the method further comprises:
randomly extracting a unit to be monitored from a plurality of units according to a random rule;
and formulating a monitoring task for the unit to be monitored to obtain the task to be processed, and determining the number of users required by the task to be processed according to the number of the unit to be monitored.
6. The method according to claim 1, wherein the step of filtering out the second user from the plurality of first users according to a preset filtering rule comprises:
acquiring all historical processing tasks of a first user, and screening out the historical processing tasks which accord with the task type of the task to be processed from all the historical processing tasks to obtain a target historical processing task;
inquiring the score value of the first user when the first user processes the target historical processing task from a pre-constructed score list; the scoring list stores scoring values which are evaluated by a system after each user processes a task;
and taking the first user with the score value higher than a preset score value as a second user.
7. The method of claim 1, wherein the pending task comprises an inspection task, and wherein the step of assigning the pending task to the second user further comprises, after the step of:
receiving the uploaded inspection data filled by the second user according to a preset standard format; the inspection data are processing data of the second user aiming at the task to be processed, and the inspection data comprise inspection items, inspection contents, deduction conditions of the inspection contents and existing problems;
and generating an inspection result according to the inspection data.
8. A task assigning apparatus, comprising:
the acquisition module is used for acquiring user information of all users and task information of the tasks to be processed; the user information and the task information are text information;
the first conversion module is used for extracting feature words from the user information of each user respectively and converting the feature words of each user into first feature vectors by using a pre-trained word vector model; the characteristic words are words representing characteristics of the user in a historical processing task process;
the second conversion module is used for extracting keywords from the task information and converting the keywords into second feature vectors by using the word vector model;
the calculating module is used for calculating the cosine distance between the first characteristic vector and the second characteristic vector of each user respectively and comparing the cosine distance obtained by calculation with a preset similarity threshold;
the screening module is used for screening a plurality of first feature vectors of which the cosine distances are greater than the preset similarity threshold value from the first feature vectors according to the comparison result to obtain a plurality of target feature vectors;
and the distribution module is used for respectively inquiring the users corresponding to the target characteristic vectors to obtain a plurality of first users, screening out second users from the first users according to a preset screening rule, and distributing the tasks to be processed to the second users.
9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor when executing the computer program implements the steps of the task assigning method according to any one of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the task assigning method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110384844.0A CN113065780B (en) | 2021-04-09 | 2021-04-09 | Task allocation method, device, storage medium and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110384844.0A CN113065780B (en) | 2021-04-09 | 2021-04-09 | Task allocation method, device, storage medium and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113065780A true CN113065780A (en) | 2021-07-02 |
| CN113065780B CN113065780B (en) | 2023-06-30 |
Family
ID=76566165
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110384844.0A Active CN113065780B (en) | 2021-04-09 | 2021-04-09 | Task allocation method, device, storage medium and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113065780B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113961726A (en) * | 2021-12-20 | 2022-01-21 | 中国人民解放军战略支援部队航天工程大学士官学校 | Command task matching method and system |
| CN113962222A (en) * | 2021-10-28 | 2022-01-21 | 重庆大学 | Availability evaluation method for demand digitization in crowdsourcing service |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109242275A (en) * | 2018-08-21 | 2019-01-18 | 平安科技（深圳）有限公司 | Method for allocating tasks, device and storage medium |
| CN109670527A (en) * | 2018-11-13 | 2019-04-23 | 平安科技（深圳）有限公司 | Acceleration recognition methods, device, computer equipment and storage medium |
| CN110222996A (en) * | 2019-06-13 | 2019-09-10 | 北京创鑫旅程网络技术有限公司 | Method for allocating tasks and device |
| CN110868500A (en) * | 2019-10-12 | 2020-03-06 | 中国平安财产保险股份有限公司 | Call communication method, device, computer equipment and storage medium |
| CN111859940A (en) * | 2019-04-23 | 2020-10-30 | 北京嘀嘀无限科技发展有限公司 | Keyword extraction method and device, electronic equipment and storage medium |
-
2021
- 2021-04-09 CN CN202110384844.0A patent/CN113065780B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109242275A (en) * | 2018-08-21 | 2019-01-18 | 平安科技（深圳）有限公司 | Method for allocating tasks, device and storage medium |
| CN109670527A (en) * | 2018-11-13 | 2019-04-23 | 平安科技（深圳）有限公司 | Acceleration recognition methods, device, computer equipment and storage medium |
| CN111859940A (en) * | 2019-04-23 | 2020-10-30 | 北京嘀嘀无限科技发展有限公司 | Keyword extraction method and device, electronic equipment and storage medium |
| CN110222996A (en) * | 2019-06-13 | 2019-09-10 | 北京创鑫旅程网络技术有限公司 | Method for allocating tasks and device |
| CN110868500A (en) * | 2019-10-12 | 2020-03-06 | 中国平安财产保险股份有限公司 | Call communication method, device, computer equipment and storage medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113962222A (en) * | 2021-10-28 | 2022-01-21 | 重庆大学 | Availability evaluation method for demand digitization in crowdsourcing service |
| CN113961726A (en) * | 2021-12-20 | 2022-01-21 | 中国人民解放军战略支援部队航天工程大学士官学校 | Command task matching method and system |
| CN113961726B (en) * | 2021-12-20 | 2022-03-01 | 中国人民解放军战略支援部队航天工程大学士官学校 | Command task matching method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113065780B (en) | 2023-06-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107122375B (en) | Image subject identification method based on image features | |
| CN109740143B (en) | Sentence distance mapping method and device based on machine learning and computer equipment | |
| CN112035636A (en) | Question-answer management method, device, equipment and storage medium of medical inquiry system | |
| CN113065780A (en) | Task allocation method, device, storage medium and computer equipment | |
| CN112507866B (en) | Chinese character vector generation method and device, computer equipment and storage medium | |
| CN112989761B (en) | Text classification method and device | |
| CN109635140B (en) | Image retrieval method based on deep learning and density peak clustering | |
| CN113642652A (en) | Method, device and equipment for generating fusion model | |
| CN114399379A (en) | Artificial intelligence-based collection behavior recognition method, device, equipment and medium | |
| CN113204968A (en) | Concept recognition method, device, equipment and storage medium of medical entity | |
| CN109685576A (en) | A kind of automatic coordination valence method, apparatus of engineering material, terminal device and storage medium | |
| CN113643046A (en) | Common situation strategy recommendation method, device, equipment and medium suitable for virtual reality | |
| CN113434763A (en) | Method, device, equipment and storage medium for generating recommendation reason of search result | |
| CN112883267A (en) | Data heat degree statistical method and device based on deep learning | |
| CN111310451B (en) | Sensitive word dictionary generating method and device, storage medium and electronic equipment | |
| CN112989054B (en) | Text processing method and device | |
| CN112989022B (en) | Intelligent virtual text selection method and device and computer equipment | |
| CN112258009B (en) | Intelligent government affair request processing method | |
| CN114756655A (en) | Data query method, device, equipment and storage medium | |
| CN109308565B (en) | Crowd performance grade identification method and device, storage medium and computer equipment | |
| CN113778681A (en) | Data processing method and device based on cloud computing and storage medium | |
| CN113342924A (en) | Answer retrieval method and device, storage medium and electronic equipment | |
| CN116663277B (en) | Method and device for prolonging service cycle based on water boiling kettle | |
| CN117171455B (en) | Grid-based high-attention event identification method and system | |
| CN115659945B (en) | Standard document similarity detection method, device and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |