Experimental path dynamic generation method
By constructing the experimental model set and full element network structure, extracting the high-activity element set and performing path search, the problems of waste of resources and inefficiency in the generation of experimental paths in the existing technology are solved, and efficient and accurate path generation is achieved.
Patent Information
- Application Number
- CN202411892749.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-20
- Publication Date
- 2025-04-11
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
When generating experimental paths, the prior art faces a huge number of path combinations, resulting in waste of computing and storage resources, and it is difficult to efficiently identify feasible paths.
By constructing experimental model sets and full-element network structures, high-activity element sets are extracted, the number of start and end points combinations of path searches is reduced, and feasible paths are quickly identified through rational judgment and path search algorithms.
It significantly improves the efficiency and accuracy of path search, reduces the consumption of computing resources, and improves the speed and quality of experimental path generation.
Smart Images
Figure CN119807662A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of intelligent technology, and in particular relates to a method for dynamically generating an experimental path. Background Art
[0002] For the data system of the prior art, the starting data is the element set and the set of associations between the elements, and the path is the data dynamically generated according to the associations between the elements. However, the experimental initial environment (starting element) and the experimental target (end point element) can theoretically be a combination of any two factors. There are many paths between the starting point and the end point of each experiment, such as a i and a j The p-node path between i and a j There are p-2 nodes connecting them. There are a total of
[0003] According to the characteristics of permutations and combinations, as n increases, the total number of potential paths will become very large. Therefore, if exhaustive element combinations are used to generate paths, the path library will become very large. In view of this, the present invention proposes an experimental path generation method that can greatly reduce the number of start and end point combinations that need to be found for the path, so that computing and storage resources can be focused on key element combinations with high activity. Summary of the invention
[0004] In order to solve the above technical problems, the present invention proposes a method for dynamically generating an experimental path to solve the above problems existing in the prior art.
[0005] To achieve the above object, the present invention provides a method for dynamically generating an experimental path, comprising:
[0006] Constructing an experimental model set, and constructing a full element set and a full element network structure based on the experimental model set;
[0007] Extracting active elements based on the full element set to obtain a high-activity element set;
[0008] A combination of the path start and end points is constructed based on the high activity element set, and a rationality judgment and path search are performed on each pair of constructed combinations to obtain a path search result.
[0009] Optionally, the process of constructing a full element set and a full element network structure based on the experimental model set includes: extracting information of main experimental elements of the experimental model set to construct a static database;
[0010] Among them, the main experimental elements include: time consumption, environmental requirements, input factors, and output factors.
[0011] Optionally, the method for obtaining a high-activity element set based on the full element set includes: measuring and screening element occurrence frequency, measuring and screening user attention, and measuring and screening factor importance.
[0012] Optionally, the process of obtaining a high-activity element set based on user attention measurement and screening includes:
[0013] Determine the starting element and the ending element;
[0014] Collect environmental information and search constraint information;
[0015] Combining the variables corresponding to the environment information and the search constraint information to construct an attribute set;
[0016] Collecting the value of the attribute set as the attribute value;
[0017] The PCA method is used to reduce the dimension of the attribute values to obtain a principal component data set;
[0018] A high activity element set is obtained based on the principal component data set.
[0019] Optionally, in the process of obtaining a high-activity element set based on the measurement and screening of factor importance, a comparative sequential analysis method is used to compare several groups of data under the same set goal, and activity values are assigned to the elements in the satisfaction level to obtain a high-activity element set.
[0020] Optionally, the principle for constructing a combination of path start and end points based on the high activity element set is:
[0021] The high-activity element sets are combined in pairs, where the elements in each combination serve as the starting and ending elements respectively.
[0022] Optionally, the process of performing rationality judgment and path search on each constructed pair of combinations to obtain path search results includes:
[0023] Make reasonable judgments on each pair of combinations based on common sense logic and professional logic;
[0024] When it is judged to be unreasonable, the current combination is marked and put into the complete solution set;
[0025] When it is judged to be reasonable, path calculation is performed on the elements of the current combination in the full-element network structure to obtain a path search result.
[0026] The present invention also provides a computer terminal device, comprising:
[0027] one or more processors;
[0028] A memory, coupled to the processor, for storing one or more programs;
[0029] When the one or more programs are executed by the one or more processors, the one or more processors implement a method for dynamically generating an experimental path.
[0030] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, a method for dynamically generating an experimental path is implemented.
[0031] Compared with the prior art, the present invention has the following advantages and technical effects:
[0032] The experimental path dynamic generation method of the present invention effectively extracts a high-activity element set by constructing an experimental model set and a full-element network structure, thereby significantly improving the efficiency and accuracy of path search. The method reduces the number of elements that need to be considered by extracting active elements, focusing on those elements that frequently appear in experiments or are highly concerned by users, thereby reducing the consumption of computing resources. In addition, through rationality judgment and path search, the method can quickly identify feasible experimental paths, avoid invalid calculations of unreasonable paths, and further improve the practicality and reliability of path generation. This method not only improves the speed of experimental path generation, but also ensures the quality of the generated path, and has important application value for experimental design and optimization in the field of intelligent technology. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0034] Figure 1 A flow chart of a method for dynamically generating an experimental path according to an embodiment of the present invention;
[0035] Figure 2 A flow chart for establishing an experimental model for an embodiment of the present invention;
[0036] Figure 3 A flowchart of constructing a full-element network structure according to an embodiment of the present invention;
[0037] Figure 4 4 is a flow chart of path search in an embodiment of the present invention. DETAILED DESCRIPTION
[0038] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.
[0039] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0040] Embodiment 1
[0041] like Figure 1 As shown, this embodiment provides a method for dynamically generating an experimental path, including the following steps: constructing an experimental model set, and constructing a full element set and a full element network structure based on the experimental model set; extracting active elements based on the full element set to obtain a high-activity element set; constructing a combination of the starting and ending points of the path based on the high-activity element set, and making rationality judgments and path searches for each pair of constructed combinations to obtain path search results.
[0042] Step S1: Full element set and full element network structure construction
[0043] 1. Establish an experimental model and determine key data through the model to collect basic data. The related data are experimental process information, input factor information, output factor information, environmental factor information, time consumption, material consumption, cost, etc. Figure 2 shown.
[0044] 2. Constructing a set of association relationships between all elements
[0045] When the output factor of one experiment is the input factor of another experiment, the two experiments form a correlation relationship between the factors.
[0046] 3. Collect experimental model data. Consult the data to extract the experimental model and establish the experimental database. According to the model defined in S1, extract the information of the main experimental elements such as time consumption, environmental requirements, input factors (elements), and output factors (elements), and import them into the system to form a static database.
[0047] 4. Establish a full element set
[0048] Traverse all experimental model data and extract all input factors (elements) and output factors (elements) as experimental element sets.
[0049] 5. Full element network structure
[0050] Through the full element set and the full element association relationship set, a predecessor and successor directed relationship can be formed in the network structure through the association relationship between elements to form a full element set. Figure 3 shown.
[0051] Step S2: Extract active elements to form a high-activity element set.
[0052] Active elements are extracted from the following channels through frequency statistical sorting, dimension reduction methods such as PCA, factor analysis methods, etc.: (1) Mining from all elements: using the frequency of elements appearing in the experiment as a measurement value; (2) Mining from user search distribution: using the distribution composed of user search frequency and other operational behaviors; (3) Others.
[0053] Furthermore, mining is done from all elements: the frequency of occurrence of elements in the experiment is used as a metric.
[0054] As a specific implementation of this embodiment, it includes:
[0055] 1. Measurement and screening of element occurrence frequency:
[0056] Total element network occurrence metric: k (ele) Value and n ele value.
[0057] The frequency of occurrence of an element in the experiment is used as a metric value: Let element ele be in k (ele) It appears in experiments, and the number of times it appears in the i-th experiment is recorded as Then the total number of occurrences is:
[0058]
[0059] For k (ele) Value and n ele The active elements obtained after sorting, screening and filtering the elements represent the active part of the entire database.
[0060] 2. Measurement and screening of user attention
[0061] When searching for related paths, you need to enter the starting element ele b and the end element ele e , collect environmental information and search constraint information at the same time, let E = (e1, e2, ..., e n ) is the environment scenario variable where the user is located, including the input and response of the previous nc searches, user portrait attributes, user historical operation information, etc., H = (h1, h2, …, h m ) are search constraint variables, including the conditions and requirements entered during the search.
[0062] Combine E and H to b ,ele e ,e1,e2,…,e n ,h1,h2,…,h m) as the attribute set, and the collected values as the attribute values, to establish a matrix, so as to enter the PCA and other dimensionality reduction methods to obtain the principal component data set of the matrix, and the principal component data set contains (ele b ,ele e ) refers to the active elements generated during the user search process. The active element set represents the active part of the element set that the user group pays attention to.
[0063] 3. Measurement and screening of factor importance
[0064] The factor analysis method can set up appropriate targets and control groups, filter and screen the factor set through comparative sequential analysis methods, so as to achieve the goal of reducing the factor set and finally extract the key factors. The steps are as follows:
[0065] (1) Create a set of elements whose activity needs to be screened;
[0066] (2) Quantitative analysis of goals, such as the achievement of search goals, satisfaction with search results, user attention, etc.;
[0067] (3) Conduct comparative analysis on two or more groups of data under one goal: such as high satisfaction and low satisfaction. The following takes satisfaction as an example.
[0068] (4) Compare and analyze the elements involved in the two groups’ satisfaction;
[0069] (5) Assign higher activity values to elements with higher satisfaction and lower activity values to elements with lower satisfaction;
[0070] Repeat the above comparative analysis, accumulate and normalize the activity values of the elements, and this is the reflection of the importance of the elements.
[0071] Step S3: Combined construction of path start and end points
[0072] The high-activity element sets are combined in pairs, and the combination principle is that each element serves as the starting point and end point element with another element respectively.
[0073] Step S4: Reasonableness judgment and construction of "complete solution set"
[0074] The rationality of each pair of such combinations is judged. For combinations without "no solution mark", the rationality of the combination of the start and end points is judged by referring to the semantic network, reasoning framework or manual confirmation. If it is judged to be unreasonable, the combination is marked as "no solution" and put into the "complete solution set". For combinations that pass the rationality judgment, proceed to step S5.
[0075] Furthermore, the process of rationality judgment and construction of the "complete solution set" includes:
[0076] (1) The basic logical framework for rationality judgment: common sense logic and professional logic.
[0077] (2) Implementation of common sense logic judgment: Calling large model APIs (such as domestic large models and chatGPT, etc.) to perform common sense judgments, such as turning stone into gold by judging that the starting element is silicon dioxide and the end element is gold, avoiding some unnecessary path calculations.
[0078] (3) Implementation of professional logic and reasoning analysis logic judgment: Collect professional knowledge, build expert systems and other systems with knowledge graph and knowledge network functions, and implement searches from a professional perspective.
[0079] (4) Logic improvement and filling: Use algorithms such as FOIL to generate logic chains.
[0080] (5) Manual analysis and review: For some frequently occurring and undetermined logic, judgments are made through manual analysis and optimization of the system model, while the logic chain is improved.
[0081] Step S5: Path search
[0082] In the full element network structure shown in S1, the path calculation of the elements of the start and end point combination is performed, and the path search between two points of the high-activity jump element combination in S4 is performed using dynamic programming algorithms such as Dijkstra or heuristic algorithms such as simulated annealing, ant colony, random walk, and machine learning algorithms. The results of searching for new paths are as follows Figure 4 shown.
[0083] If no new path is found, it will be marked as "no new solution". If there is no available path for this combination, it will be marked as "no solution". Then put it into the "complete solution set". As a specific implementation of this embodiment, it includes:
[0084] (1) Determine the search method. Different search methods are planned, such as depth-first search DFS, breadth-first search BFS, dijkstra, and dynamic programming. Select one of them for search. Take dynamic programming combined with heuristic algorithm as an example.
[0085] (2) Search along a path planned by the algorithm, starting from the starting element and ending at the ending element. At each search fork, a data set VSet is generated. i, put all nodes that can be searched later into the data set, indicating the number of nodes that can be selected at the fork. When the target element is found or the search times exceed the preset threshold (or the algorithm judgment of the heuristic algorithm) and the target is still not found, the path is returned, and whether the search is successful is marked. The path is:
[0086] p i =(p i1 ,p i2 ,…,p im , mark)
[0087] Among them, p i1 ,p i2 ,…,p im is a path set, and mark represents the search result.
[0088] (3) Return to the search set VSet that still has the remaining unselected nodes k , select a new search node to continue searching, and continue (2) until all nodes are added to the path, forming a path set, which is the planning strategy. Thus, the strategy (path set) is formed.
[0089] P=(p1,p2,…,p n )
[0090] During the search process, when the number of available nodes is large and the search range is wide, heuristic search methods can be used in conjunction with it, such as simulated annealing search, evolutionary algorithm, ant colony algorithm, etc.
[0091] (4) When the algorithm iteration meets the preset conditions of the algorithm, the search process is exited.
[0092] (5) Traverse the search results and mark those that do not find new paths as no new solutions. Mark the element combinations that cannot find paths as no solutions. This is used to strengthen the knowledge base of the above-mentioned "rational logical judgment".
[0093] (6) Generate and store new paths: Generate new path data for the newly searched path and store it in the path database.
[0094] Embodiment 2
[0095] Take the synthetic experimental path from "iron oxide" to "LK-99" as an example:
[0096] Its users work in the front-end including:
[0097] Step 1: First, enter the starting point as "iron oxide" and the end point as "LK-99" in the search box and click Search.
[0098] Step 2: The database is matched by paths to obtain the synthesis path.
[0099] Its staff working behind the scenes include:
[0100] Step 1: Build a data set: including the experiment name, experimental elements, experimental conditions, etc.
[0101] Step 2: Experimental elements are the public data of "experiments". Different experiments can reference the same elements, thus building the relationship between elements with experiments as the bridge.
[0102] Step 3: Build a path search algorithm through association relationships to generate a path set and return results when the user searches.
[0103] Embodiment 3
[0104] This embodiment also provides a computer terminal device, including:
[0105] one or more processors;
[0106] A memory, coupled to the processor, for storing one or more programs;
[0107] When the one or more programs are executed by the one or more processors, the one or more processors implement a method for dynamically generating an experimental path.
[0108] This embodiment also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, a method for dynamically generating an experimental path is implemented.
[0109] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A method for dynamically generating an experimental path, characterized in that: The following steps are involved: Constructing an experimental model set, and constructing a full element set and a full element network structure based on the experimental model set; Extracting active elements based on the full element set to obtain a high-activity element set; A combination of the path start and end points is constructed based on the high activity element set, and a rationality judgment and path search are performed on each pair of constructed combinations to obtain a path search result.
2. The method according to claim 1, characterized in that The process of constructing a full element set and a full element network structure based on the experimental model set includes: extracting information of main experimental elements of the experimental model set to construct a static database; Among them, the main experimental elements include: time consumption, environmental requirements, input factors, and output factors.
3. The method according to claim 1, characterized in that: The method for obtaining a high-activity element set based on the full element set includes: measuring and screening element occurrence frequency, measuring and screening user attention, and measuring and screening factor importance.
4. The method according to claim 3, characterized in that The process of measuring and filtering to obtain a set of highly active elements based on user attention includes: Determine the starting element and the ending element; Collect environmental information and search constraint information; Combining the variables corresponding to the environment information and the search constraint information to construct an attribute set; Collecting the value of the attribute set as the attribute value; The PCA method is used to reduce the dimension of the attribute values to obtain a principal component data set; A high activity element set is obtained based on the principal component data set.
5. The method according to claim 3, characterized in that: In the process of obtaining a high-activity element set based on the measurement and screening of factor importance, several groups of data under the same set goal are compared through the comparative sequential analysis method, and the elements in the satisfaction are assigned activity values to obtain a high-activity element set.
6. The method according to claim 1, characterized in that The principle of constructing the combination of the path start and end points based on the high activity element set is: The high-activity element sets are combined in pairs, where the elements in each combination serve as the starting and ending elements respectively.
7. The method according to claim 6, characterized in that The process of making reasonable judgment and path search for each pair of constructed combinations to obtain path search results includes: Make reasonable judgments on each pair of combinations based on common sense logic and professional logic; When it is judged to be unreasonable, the current combination is marked and put into the complete solution set; When it is judged to be reasonable, path calculation is performed on the elements of the current combination in the full-element network structure to obtain a path search result.
8. A computer terminal device, characterized in that: include: one or more processors; A memory, coupled to the processor, for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method for dynamically generating an experimental path as described in any one of claims 1-7.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for dynamically generating an experimental path as described in any one of claims 1 to 7 is implemented.