Decision ability evaluation method and device, electronic equipment and storage medium

Abstract

The invention provides a decision ability evaluation method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a group discussion text which is formed according to the participation discussion of a to-be-evaluated participant; based on the group discussion text, a participant influence graph corresponding to the to-be-evaluated participant is obtained, and the participant influence graph comprises graph nodes and graph edges corresponding to behaviors formed by the to-be-evaluated participant in the group discussion process; and based on the participant influence map, performing capability evaluation on the decision-making capability of the to-be-evaluated participant under a preset capability evaluation index. The decision-making ability of the individual ability in group discussion can be automatically and objectively evaluated, and the method is of great significance in improving the quality of group discussion and optimizing the decision-making process.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a decision-making ability evaluation method, device, electronic equipment and storage medium. Background Art

[0002] In the field of public administration and in the internal operation of enterprises, group discussion has become a widely adopted decision-making model in the face of complex and far-reaching decision-making tasks. Compared with individual decision-making, group discussion has shown many advantages. However, despite the many advantages of group discussion, some problems that cannot be ignored have also been exposed in the actual operation process, such as groupthink, production blocking, polarization, etc. In view of these problems in group discussion, the analysis of individual ability in group discussion is particularly important.

[0003] However, traditional individual ability analysis methods often rely on manual evaluation, which has shortcomings such as strong subjectivity, time-consuming and labor-intensive.

[0004] Therefore, developing a decision-making ability assessment method that can automatically and objectively evaluate individual abilities in group discussions has become a current research hotspot. Summary of the invention

[0005] The present invention provides a decision-making ability assessment method, device, electronic device and storage medium, which realize the decision-making ability that can automatically and objectively assess the individual ability in group discussion, which is of great significance for improving the quality of group discussion and optimizing the decision-making process.

[0006] The present invention provides a decision-making ability assessment method, the method comprising: obtaining a group discussion text, wherein the group discussion text is formed according to the participation discussion of participants to be evaluated; based on the group discussion text, obtaining a participant influence map corresponding to the participants to be evaluated, wherein the participant influence map includes map nodes and map edges corresponding to the behaviors formed by the participants to be evaluated in the process of participating in the group discussion; based on the participant influence map, performing ability assessment on the decision-making ability of the participants to be evaluated under preset ability assessment indicators.

[0007] According to a decision-making ability assessment method provided by the present invention, when the participant influence map includes a preset sub-graph structure, the decision-making ability of the participant to be evaluated under the preset ability evaluation index is evaluated based on the participant influence map, specifically comprising: extracting the preset sub-graph structure obtained in the participant influence map based on the participant influence map; calling a preset standard sub-graph structure, wherein the standard sub-graph structure is a sub-graph structure corresponding to a participant with decision-making ability under the preset ability evaluation index; based on the preset sub-graph structure and the standard sub-graph structure, the decision-making ability of the participant to be evaluated under the preset ability evaluation index is evaluated.

[0008] According to a decision-making ability assessment method provided by the present invention, the decision-making ability of the participant to be evaluated under the preset ability evaluation index is evaluated based on the preset subgraph structure and the standard subgraph structure, specifically including: when the preset subgraph structure matches the standard subgraph structure, determining that the participant to be evaluated has the decision-making ability under the preset ability evaluation index; when the preset subgraph structure does not match the standard subgraph structure, determining that the participant to be evaluated does not have the decision-making ability under the preset ability evaluation index.

[0009] According to a decision-making ability assessment method provided by the present invention, when the participant influence map does not include a preset subgraph structure, before the decision-making ability of the participant to be evaluated under the preset ability evaluation index is evaluated based on the participant influence map, the method also includes: constructing a first prompt text, wherein the first prompt text is used to represent the target node information extracted from the participant influence map for evaluating the decision-making ability under the preset ability evaluation index; the decision-making ability of the participant to be evaluated under the preset ability evaluation index based on the participant influence map specifically includes: calling a generative pre-trained transformer model; inputting the participant influence map and the first prompt text into the generative pre-trained transformer model to obtain the target node information output by the generative pre-trained transformer model; and based on the target node information, evaluating the decision-making ability of the participant to be evaluated under the preset ability evaluation index.

[0010] According to a decision-making ability assessment method provided by the present invention, the decision-making ability of the participant to be evaluated under the preset ability evaluation index is evaluated based on the target node information, specifically including: when the target node information meets the preset node information requirements, determining that the participant to be evaluated has the decision-making ability under the preset ability evaluation index; when the target node information does not meet the preset node information requirements, determining that the participant to be evaluated does not have the decision-making ability under the preset ability evaluation index, wherein the preset node information requirements are node information requirements corresponding to the decision-making ability of the participant to be evaluated under the preset ability evaluation index.

[0011] According to a decision-making ability assessment method provided by the present invention, when the participant influence map does not include a preset subgraph structure, the decision-making ability of the participant to be evaluated under the preset ability evaluation index is evaluated based on the participant influence map, specifically including: based on the participant influence map, determining the target node information in the participant influence map, wherein the target node information is used to characterize the node information in the participant influence map for evaluating the decision-making ability under the preset ability evaluation index; when the target node information meets the preset node information requirements, determining that the participant to be evaluated has the decision-making ability under the preset ability evaluation index; when the target node information does not meet the preset node information requirements, determining that the participant to be evaluated does not have the decision-making ability under the preset ability evaluation index, wherein the preset node information requirements are the node information requirements corresponding to the decision-making ability of the participant to be evaluated under the preset ability evaluation index.

[0012] According to a decision-making ability assessment method provided by the present invention, the method of obtaining a participant influence map corresponding to the participant to be evaluated based on the group discussion text specifically includes: based on the group discussion text, obtaining a map node with a preset schematic structure corresponding to the group discussion text, and a map edge with a preset schematic structure corresponding to the group discussion text; based on the map node with a preset schematic structure, and the map edge with a preset schematic structure, obtaining a participant influence map corresponding to the participant to be evaluated.

[0013] According to a decision-making ability assessment method provided by the present invention, based on the group discussion text, a graph node with a preset schematic structure corresponding to the group discussion text is obtained, which specifically includes: constructing a second prompt text, wherein the second prompt text is used to characterize and output a graph node with a preset schematic structure corresponding to the group discussion text, wherein the graph node includes at least a chance node, a decision node and a utility node; calling a generative pre-trained transformer model; inputting the group discussion text and the second prompt text into the generative pre-trained transformer model, and obtaining the graph node with a preset schematic structure corresponding to the group discussion text output by the generative pre-trained transformer model.

[0014] According to a decision-making ability assessment method provided by the present invention, the method of obtaining a graph edge with a preset schematic structure corresponding to the group discussion text based on the group discussion text specifically includes: constructing a third prompt text, wherein the third prompt text is used to characterize and output a graph edge with a preset schematic structure corresponding to the group discussion text; calling a generative pre-trained transformer model; inputting the group discussion text, the graph node with a preset schematic structure, and the third prompt text into the generative pre-trained transformer model to obtain the graph edge with a preset schematic structure corresponding to the group discussion text output by the generative pre-trained transformer model.

[0015] The present invention also provides a decision-making ability assessment device, which includes: an acquisition module, used to acquire a group discussion text, wherein the group discussion text is formed according to the participation discussion of the participants to be evaluated; a processing module, used to obtain a participant influence map corresponding to the participants to be evaluated based on the group discussion text, wherein the participant influence map includes map nodes and map edges corresponding to the behaviors formed by the participants to be evaluated in the process of participating in the group discussion; an evaluation module, used to evaluate the decision-making ability of the participants to be evaluated under preset ability evaluation indicators based on the participant influence map.

[0016] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, any one of the decision-making capability assessment methods described above is implemented.

[0017] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the decision-making capability assessment method as described above is implemented.

[0018] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the decision-making ability assessment method described above is implemented.

[0019] The decision-making ability evaluation method, device, electronic device and storage medium provided by the present invention obtain a group discussion text, wherein the group discussion text is formed according to the participation discussion of the participants to be evaluated; based on the group discussion text, a participant influence map corresponding to the participants to be evaluated is obtained, wherein the participant influence map includes map nodes and map edges corresponding to the behaviors formed by the participants to be evaluated in the process of participating in the group discussion; based on the participant influence map, the decision-making ability of the participants to be evaluated under the preset ability evaluation indicators is evaluated. The decision-making ability that can automatically and objectively evaluate the individual ability in the group discussion is realized, which is of great significance for improving the quality of group discussion and optimizing the decision-making process. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 It is a flow chart of the decision-making ability assessment method provided by the present invention.

[0022] Figure 2 It is one of the flowchart diagrams provided by the present invention for evaluating the decision-making ability of the participant to be evaluated under the preset ability evaluation index based on the participant influence map.

[0023] Figure 3 This is the second flow chart of the present invention for evaluating the decision-making ability of the participant to be evaluated under the preset ability evaluation index based on the participant influence map.

[0024] Figure 4 The present invention provides a flow chart of obtaining a graph node with a preset diagram structure corresponding to the group discussion text based on the group discussion text.

[0025] Figure 5 The present invention provides a flow chart of obtaining a graph edge with a preset schematic structure corresponding to the group discussion text based on the group discussion text.

[0026] Figure 6 It is a structural schematic diagram of the decision-making ability evaluation device provided by the present invention.

[0027] Figure 7 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] The decision-making ability assessment method provided by the present invention automatically extracts influence diagrams from knowledge bases and group discussion texts based on a large language model, classifies nodes and directed edges in the diagrams, and organizes six aspects of ability evaluation indicators based on the adjusted Howard (2000) theory, completing the automatic assessment process of decision analysis and innovation inspiration ability from text input to output.

[0030] Figure 1 It is a flow chart of the decision-making ability assessment method provided by the present invention.

[0031] The following will be combined Figure 1 The process of the decision-making ability assessment method provided by the present invention is described.

[0032] In an exemplary embodiment of the present invention, Figure 1 It can be seen that the decision-making ability assessment method may include steps 110 to 130, and each step will be introduced below.

[0033] In step 110, a group discussion text is obtained, wherein the group discussion text is formed according to the discussion participated in by the participants to be evaluated.

[0034] In one embodiment, a group discussion text may be obtained, wherein the group discussion text is formed based on the participation of the to-be-evaluated participants in the group discussion process.

[0035] In step 120, based on the group discussion text, a participant influence map corresponding to the participant to be evaluated is obtained, wherein the participant influence map includes map nodes and map edges corresponding to the behaviors formed by the participant to be evaluated during the group discussion.

[0036] In step 130, based on the participant influence map, the decision-making ability of the participant to be evaluated is evaluated under the preset ability evaluation index.

[0037] In another embodiment, a participant influence map corresponding to the participant to be evaluated can be obtained from the group discussion text, wherein the participant influence map can include map nodes and map edges corresponding to the behaviors formed by the participant to be evaluated during the group discussion. It can be understood that the participant influence map is an influence map of each participant obtained based on the group discussion text.

[0038] In another embodiment, based on the prompts of the large language model, information and its directed relationships can be extracted from the knowledge base and group discussion text respectively to generate the influence diagram of the participants (corresponding to the participant influence map). Specifically, the large language model can automatically identify three types of key information: information nodes representing uncertainty (opportunity nodes), nodes representing decision plans (decision nodes), and nodes representing decision goals and values ​​(utility nodes); at the same time, the specific value of each node in the text is extracted. The large language model can be used to extract the causal relationship between nodes from the knowledge base, extract all directed relationships from the group discussion, and automatically determine the conditional probabilities on these relationships, so that the graph edges can be obtained.

[0039] In yet another embodiment, the decision-making ability of the participant to be evaluated under preset ability evaluation indicators may be evaluated based on the participant influence map.

[0040] Among them, the decision-making ability under the preset ability evaluation index can include the decision-making ability under the energy evaluation index under the dimension in Howard (2000) theory, which can include the hypothesis construction ability and target scope definition ability in terms of framework; the target decomposition ability, value weighing ability, and preference expression in terms of value; the information updating ability, information association ability, and content gain in terms of information; the solution association ability, value-based decision-making ability, and consequence prediction ability in terms of options; the feature transfer ability in terms of logic; critical thinking and leadership in terms of teamwork, etc. In this embodiment, the decision-making ability under the preset ability evaluation index is not specifically limited, and it can be adjusted accordingly according to actual conditions.

[0041] The decision-making ability evaluation method provided by the present invention obtains a group discussion text, wherein the group discussion text is formed according to the participation discussion of the participants to be evaluated; based on the group discussion text, a participant influence map corresponding to the participants to be evaluated is obtained, wherein the participant influence map includes map nodes and map edges corresponding to the behaviors formed by the participants to be evaluated in the process of participating in the group discussion; based on the participant influence map, the decision-making ability of the participants to be evaluated under the preset ability evaluation index is evaluated. The decision-making ability that can automatically and objectively evaluate the individual ability in the group discussion is realized, which is of great significance for improving the quality of group discussion and optimizing the decision-making process.

[0042] Figure 2It is one of the flowchart diagrams provided by the present invention for evaluating the decision-making ability of the participant to be evaluated under the preset ability evaluation index based on the participant influence map.

[0043] The following will be combined Figure 2 The process of evaluating the decision-making ability of the participant to be evaluated under the preset ability evaluation index based on the participant influence map provided by the present invention is explained.

[0044] In an exemplary embodiment of the present invention, when the participant influence graph includes a preset subgraph structure, Figure 2 It can be seen that based on the participant influence map, the decision-making ability of the participant to be evaluated under the preset ability evaluation index is evaluated, which can include steps 210 to 230, and each step will be introduced below.

[0045] In step 210, based on the participant influence graph, a preset subgraph structure in the participant influence graph is extracted.

[0046] In one embodiment, a preset subgraph structure in the participant influence map can be extracted based on the participant influence map. In the application process, the abstract attributes associated with each node in the participant influence map can be extracted so that the specific information can be classified into a higher-level concept, and the abstract attributes are added to the participant influence map. For nodes with a specific subgraph structure in the participant influence map, normal distribution or beta distribution is used for assignment.

[0047] In another embodiment, the nodes in the participant influence map can be abstracted to obtain an abstracted participant influence map, and a preset subgraph structure in the participant influence map can be extracted based on the abstracted participant influence map. Each graph node in the abstracted participant influence map is divided into a specific information node and an abstract information node.

[0048] In step 220, a preset standard subgraph structure is called, wherein the standard subgraph structure is a subgraph structure corresponding to a participant having decision-making ability under a preset ability evaluation index.

[0049] In step 230, based on the preset subgraph structure and the standard subgraph structure, the decision-making ability of the participant to be evaluated is evaluated under the preset ability evaluation index.

[0050] In another embodiment, a pre-set standard subgraph structure may be called, wherein the standard subgraph structure is a subgraph structure corresponding to a participant having decision-making ability under a preset ability evaluation index. It should be noted that the standard subgraph structure is determined based on the subgraph structure defined on the reference answer influence graph, and the reference answer influence graph is a graph generated by extracting a knowledge base.

[0051] It is understandable that different standard subgraph structures may correspond to subgraph structures corresponding to participants with decision-making capabilities under different preset capability evaluation indicators. In another example, the capabilities of the participants are judged by defining the subgraph structure on the reference answer influence diagram and identifying the corresponding structure in the group discussion text. The capabilities involving specific subgraph structures include goal decomposition capability, value weighing capability, information updating capability, information association capability, solution association capability, value-based decision-making capability, and consequence prediction capability. It is understandable that the decision-making capabilities under the preset capability evaluation indicators described in steps 220 and 230 can be considered as the aforementioned capabilities.

[0052] In another embodiment, the decision-making ability of the evaluation participants under the preset ability evaluation indicators can be evaluated based on the preset subgraph structure and the standard subgraph structure, thereby achieving the ability evaluation of the decision-making ability of the evaluation participants under the preset ability evaluation indicators, which is of great significance for improving the quality of group discussions and optimizing the decision-making process.

[0053] In another exemplary embodiment of the present invention, continuing with the above Figure 2 The above embodiment is used as an example for explanation, wherein, based on the preset subgraph structure and the standard subgraph structure, the decision-making ability of the evaluation participant under the preset ability evaluation index is evaluated, which can be implemented in the following manner: In the case where the preset subgraph structure matches the standard subgraph structure, determining that the participant to be evaluated has decision-making ability under the preset ability evaluation index; In the case where the preset subgraph structure does not match the standard subgraph structure, it is determined that the participant to be evaluated does not have the decision-making ability under the preset ability evaluation index.

[0054] The following will explain the process of judging participants' goal decomposition ability, value weighing ability, information updating ability, information association ability, solution association ability, value-based decision-making ability, and consequence prediction ability.

[0055] In another embodiment, the ability to decompose a goal refers to the ability to decompose a goal into an ultimate goal and a means goal, and to make decisions based on the ultimate goal. In the reference answer influence diagram (corresponding to the standard subgraph structure), the subgraph structure corresponding to this ability is: the decision points to the (causal) means goal, the means goal points to the (causal) ultimate goal, and there is no direct connection between the decision and the ultimate goal. For each such subgraph structure, if the participant mentions the decision and the ultimate goal at the same time in the same speech, that is, the preset subgraph structure reflects this information, in other words, when the preset subgraph structure matches the standard subgraph structure, it can be judged that the participant to be evaluated has demonstrated the ability to decompose the goal.

[0056] In another embodiment, the value balancing ability refers to making trade-offs between multiple values ​​or goals and tapping into hidden values. In the reference answer influence diagram (corresponding to the standard subgraph structure), the subgraph structure corresponding to this ability is: the decision points to (causally) at least one utility, and is not directly connected to at least one utility but d-connected. For each such subgraph structure, if a participant mentions two utilities at the same time in the same speech, that is, the preset subgraph structure reflects this information, in other words, when the preset subgraph structure matches the standard subgraph structure, it can be judged that the participant to be evaluated reflects the value balancing ability.

[0057] In another embodiment, the information updating capability refers to updating the knowledge of unknown information based on the current known information. In the reference answer influence diagram (corresponding to the standard subgraph structure), the subgraph structure corresponding to this capability is: unknown information points to (causal) known information, and known information points to (evidence update) unknown information. For each such subgraph structure, if the participant mentions known information and unknown information at the same time in the same speech, and correctly expresses the positive and negative directions of the relationship between the information (judged by the large model), that is, the preset subgraph structure reflects this information. In other words, when the preset subgraph structure matches the standard subgraph structure, it can be judged that the participant to be evaluated reflects the information updating capability.

[0058] A special type of subgraph structure is that the unknown information is not affected by other information, but only affects one or more currently known information, which is also only affected by the unknown information. At this time, the unknown information is modeled as a normal distribution (continuous) or a Beta distribution (discrete), and the known information is correspondingly modeled as a normal distribution (continuous) or a binomial distribution (discrete). The parameters are determined by the large model so that the posterior distribution of the unknown information can be written. At this time, if the participants can quantitatively discuss the posterior distribution of the unknown information and it conforms to the actual situation (judged by the large model), it shows that their information updating ability is very strong.

[0059] In another embodiment, the information association ability refers to the ability to associate existing information with information that has not yet been considered. In the reference answer influence diagram (corresponding to the standard subgraph structure), the subgraph structure corresponding to this ability is: at least one information or decision points to (causally) two pieces of information at the same time, and the two pieces of information are not directly connected. For each such subgraph structure, if a participant mentions two pieces of information successively in the same speech, and the second piece of information has not appeared in the group discussion text before, that is, the preset subgraph structure reflects this information. In other words, when the preset subgraph structure matches the standard subgraph structure, it can be judged that the participant to be evaluated has demonstrated the ability to associate information.

[0060] In another embodiment, the ability to associate solutions refers to associating existing solutions with solutions that have not yet been considered. In the reference answer influence diagram, the subgraph structure corresponding to this ability is: two decisions simultaneously point to (causally) at least one information or utility, and the two decisions are not directly connected. For each such subgraph structure, if a participant mentions two decisions successively in the same speech, and the second decision has not appeared in the group discussion text before, that is, the preset subgraph structure reflects this information. In other words, when the preset subgraph structure matches the standard subgraph structure, it can be judged that the participant to be evaluated has demonstrated the ability to associate solutions.

[0061] In another embodiment, the value-based decision-making ability refers to refining and deepening the decision-making plan based on value. In the reference answer influence diagram corresponding to the standard subgraph structure, the subgraph structure corresponding to this ability is: information and decision both point to (causal) utility, and information points to (information) decision. For each such subgraph structure, if the participant mentions information and decision at the same time in the same speech, that is, the preset subgraph structure reflects this information, in other words, when the preset subgraph structure matches the standard subgraph structure, it can be judged that the participant to be evaluated has demonstrated the value-based decision-making ability.

[0062] In another embodiment, the ability to predict consequences refers to predicting and evaluating the possible consequences of different decisions. In the reference answer influence diagram (corresponding to the standard subgraph structure), the subgraph structure corresponding to this ability is: decision-making points to (causal) information, and information points to (information) decision-making. For each such subgraph structure, if the participant mentions both information and decision-making in the same speech, and correctly expresses the positive and negative direction of the relationship between information and decision-making (judged by the large model), that is, the preset subgraph structure reflects this information. In other words, when the preset subgraph structure matches the standard subgraph structure, it can be judged that the participant to be evaluated has demonstrated the ability to predict consequences.

[0063] It should be noted that the decision, means target, ultimate target and other nodes mentioned above are obtained by redivision of the utility nodes. They can be implemented based on a large model, and will not be described in detail in this embodiment.

[0064] Figure 3 This is the second flow chart of the present invention for evaluating the decision-making ability of the participant to be evaluated under the preset ability evaluation index based on the participant influence map.

[0065] The following will be combined Figure 3 Another process of evaluating the decision-making ability of the participant to be evaluated under preset ability evaluation indicators based on the participant influence map is described.

[0066] In an exemplary embodiment of the present invention, when the participant influence graph does not include a preset subgraph structure, Figure 3 It can be seen that, based on the participant influence map, the ability evaluation of the decision-making ability of the participant to be evaluated under the preset ability evaluation index may include steps 310 to 340, and each step will be introduced below.

[0067] In step 310, a first prompt text is constructed, wherein the first prompt text is used to represent target node information extracted from the participant influence map for evaluating decision-making capabilities under preset capability evaluation indicators.

[0068] In step 320, the generative pre-trained transformer model is called.

[0069] In step 330, the participant influence map and the first prompt text are input into the generative pre-trained transformer model to obtain the target node information output by the generative pre-trained transformer model.

[0070] In one embodiment, a first prompt text (prompt) can be constructed, wherein the first prompt text is used to characterize the target node information extracted from the participant influence map for evaluating the decision-making ability under the preset ability evaluation index. In the process of evaluating the hypothesis construction ability, a large model (generative pre-trained transformer model) can be used to analyze which assumptions the participants have made, whether these assumptions cover the key information nodes necessary to solve the decision-making problem (determined in advance based on domain knowledge), and whether the values ​​of the assumptions are reasonable. It is understandable that the information mentioned above can be considered as target node information.

[0071] In step 340, based on the target node information, the decision-making ability of the participant to be evaluated is evaluated under the preset ability evaluation index.

[0072] In yet another embodiment, the decision-making ability of the participant to be evaluated may be evaluated based on the target node information under preset ability evaluation indicators.

[0073] In another exemplary embodiment of the present invention, continuing with the above Figure 3 The above embodiment is used as an example for explanation, wherein, based on the target node information, the decision-making ability of the participant to be evaluated under the preset ability evaluation index is evaluated (corresponding to step 340) and can be implemented in the following manner: When the target node information meets the preset node information requirements, it is determined that the participant to be evaluated has the decision-making ability under the preset ability evaluation index; When the target node information does not meet the preset node information requirements, it is determined that the participant to be evaluated does not have the decision-making ability under the preset ability evaluation index, where: The preset node information requirements are the node information requirements corresponding to the decision-making capabilities of the participant to be evaluated under the preset capability evaluation indicators.

[0074] In one embodiment, the evaluation of hypothesis building capability is continued as an example. After analyzing the assumptions made by the participants through a large model (generative pre-trained transformer model), whether these assumptions cover the key information nodes necessary for solving decision problems (determined in advance based on domain knowledge), and whether the values ​​of the assumptions are reasonable, the hypothesis building capability can be measured by the proportion of the number of reasonable assumptions to the key information nodes. The preset node information requires that the proportion of the number of reasonable assumptions to the key information nodes meets the preset value.

[0075] In another embodiment, the above scheme can also be used to evaluate the target scope definition ability. The target scope definition ability is used to determine whether the participants can clearly define the scope and scale of the decision-making goals. The evaluation method is: extract the utility nodes mentioned by the participants and the statements about their value range and variance through the large model to determine whether the value range is reasonable. The target scope definition ability is measured by the ratio of the number of targets with reasonable value ranges to the number of utility nodes.

[0076] In order to further introduce the decision-making ability evaluation method provided by the present invention, another process of evaluating the decision-making ability of the participant to be evaluated under the preset ability evaluation index based on the participant influence map when the participant influence map does not include the preset sub-graph structure will be described below.

[0077] In another exemplary embodiment of the present invention, when the participant influence map does not include a preset sub-map structure, the decision-making ability of the participant to be evaluated under the preset ability evaluation index is evaluated based on the participant influence map, specifically including: Based on the participant influence map, determine the target node information in the participant influence map, wherein the target node information is used to represent the node information in the participant influence map for evaluating the decision-making ability under the preset ability evaluation index; When the target node information meets the preset node information requirements, it is determined that the participant to be evaluated has the decision-making ability under the preset ability evaluation index; When the target node information does not meet the preset node information requirements, it is determined that the participant to be evaluated does not have the decision-making ability under the preset ability evaluation index, where: The preset node information requirements are the node information requirements corresponding to the decision-making capabilities of the participant to be evaluated under the preset capability evaluation indicators.

[0078] In one embodiment, the target node information in the participant influence map can be determined directly based on the node information contained in the participant influence map, wherein the target node information can be used to characterize the node information in the participant influence map used to evaluate the decision-making ability under the preset ability evaluation index. Further, when the target node information meets the preset node information requirements, it can be determined that the participant to be evaluated has the decision-making ability under the preset ability evaluation index; when the target node information does not meet the preset node information requirements, it is determined that the participant to be evaluated does not have the decision-making ability under the preset ability evaluation index. Among them, the preset node information requirements can be considered as the node information requirements corresponding to the decision-making ability of the participant to be evaluated under the preset ability evaluation index.

[0079] In the application process, the feature transfer ability, critical thinking, and leadership can be tested according to this scheme. In one example, the feature transfer ability is used to determine whether the participant can discover universal features from a specific phenomenon (i.e., discover abstract attributes from specific information) and generalize the feature to other phenomena (other specific information with the same abstract attributes). If it is determined based on the participant influence map that the participant mentioned two specific information separated by d in the reference answer influence map in the corpus, and the "abstract attributes" of the two specific information are the same, it can be judged that the feature transfer ability is reflected, and each mention is scored as 1 point.

[0080] In another embodiment, critical thinking evaluates whether participants can critically analyze existing viewpoints and propose information that is semantically different from previous discussions. Critical thinking is evaluated by the average semantic distance between each node mentioned by the participant and all previous nodes. The semantic distance is calculated based on the Wiki Chinese pre-trained word vectors. The aforementioned content can be obtained from the participant's influence map, corresponding to the target node information.

[0081] In another embodiment, leadership evaluation is whether participants can provide discussion direction, organize discussion process, discover and resolve conflicts in group discussions, etc. Such behaviors are generally considered to reflect leadership. Leadership is judged by the big model from the speech content of each participant in the group discussion, and the big model scores. The aforementioned content can be obtained from the participant influence map, corresponding to the target node information.

[0082] In another exemplary embodiment of the present invention, continuing the above Figure 1 The above embodiment is used as an example for explanation, wherein, based on the group discussion text, obtaining the participant influence map corresponding to the participant to be evaluated (corresponding to step 120) can be implemented in the following manner: Based on the group discussion text, a graph node with a preset schema structure corresponding to the group discussion text and a graph edge with a preset schema structure corresponding to the group discussion text are obtained respectively; Based on the graph nodes with the preset schema structure and the graph edges with the preset schema structure, a participant influence graph corresponding to the participant to be evaluated is obtained.

[0083] In one embodiment, based on a large model, such as chat-GPT, and a group discussion text, graph nodes with a preset schema structure corresponding to the group discussion text and graph edges with a preset schema structure corresponding to the group discussion text can be obtained respectively; then, based on the graph nodes with the preset schema structure and the graph edges with the preset schema structure, a participant influence graph corresponding to the participant to be evaluated is obtained.

[0084] It should be noted that the preset graph structure can be determined according to actual conditions and is not specifically limited in the present embodiment. For example, the large language model is required to output a structured node list that conforms to the element graph, and the large language model is required to output the conditional probability dependencies in the nodes and output the edges in the form of a list, where each element in the list is a dictionary that conforms to the element graph requirements.

[0085] Figure 4 The present invention provides a flow chart of obtaining a graph node with a preset diagram structure corresponding to the group discussion text based on the group discussion text.

[0086] The following will be combined Figure 4 The process of obtaining a graph node with a preset schema structure corresponding to the group discussion text based on the group discussion text is explained.

[0087] In an exemplary embodiment of the present invention, Figure 4 It can be seen that obtaining a graph node with a preset schema structure corresponding to the group discussion text based on the group discussion text may include steps 410 to 430, and each step will be introduced below.

[0088] In step 410, a second prompt text is constructed, wherein the second prompt text is used to characterize and output a graph node having a preset graph structure corresponding to the group discussion text, wherein the graph node includes at least a chance node, a decision node, and a utility node.

[0089] In step 420, the generative pre-trained transformer model is called.

[0090] In step 430, the group discussion text and the second prompt text are input into the generative pre-trained transformer model to obtain a graph node with a preset schema structure corresponding to the group discussion text output by the generative pre-trained transformer model.

[0091] In one embodiment, variables can be extracted from unstructured text (corresponding to group discussion text), and these variables will be represented as nodes in the influence diagram. Input unstructured text, and require the large language model (corresponding to the generative pre-trained transformer model) to output a structured node list that conforms to the element diagram. Nodes include three types: opportunity nodes, decision nodes, and utility nodes. The node extraction method uses the language understanding and conditional generation capabilities of the large language model to guide the model to identify variables in the text by designing prompts (corresponding to the second prompt text) (prompt) and determine their types according to the context. First, prepare the input prompt through the prompt template, and combine the input unstructured text with the pre-set prompt template. The prompt contains clear guidance on node extraction, for example, requiring the model to first extract utility nodes, then extract decision nodes, and finally extract opportunity nodes. The prompt also makes clear requirements on the output format, representing the extracted nodes in the form of a dictionary, and storing the node list in JSON format. The record includes the name, type, and possible values ​​of the node. It can be understood that the description in the previous text can be considered as a preset diagram structure.

[0092] Figure 5 The present invention provides a flow chart of obtaining a graph edge with a preset schematic structure corresponding to the group discussion text based on the group discussion text.

[0093] The following will be combined Figure 5 The process of obtaining a graph edge having a preset schematic structure corresponding to the group discussion text based on the group discussion text is described.

[0094] In an exemplary embodiment of the present invention, Figure 5 It can be seen that, based on the group discussion text, obtaining a graph edge with a preset schematic structure corresponding to the group discussion text may include steps 510 to 530, and each step will be introduced below.

[0095] In step 510, a third prompt text is constructed, wherein the third prompt text is used to represent and output a graph edge having a preset schematic structure corresponding to the group discussion text.

[0096] In step 520, the generative pre-trained transformer model is called.

[0097] In step 530, the group discussion text, the graph nodes with a preset schematic structure, and the third prompt text are input into the generative pre-trained transformer model to obtain the graph edges with a preset schematic structure corresponding to the group discussion text output by the generative pre-trained transformer model.

[0098] In one embodiment, dependencies between nodes can be identified and represented as edges. Edges represent potential causal relationships or statistical correlations between nodes. Edge extraction depends on the identification results of the nodes. During application, the model (corresponding to the generative pre-trained transformer model) uses context analysis and generates probabilistic dependencies based on the extracted nodes. Each edge generally represents the probability distribution of a node given the values ​​of other nodes.

[0099] In extraction, given the input text and nodes, the large language model extracts the conditional probability dependencies in the nodes and outputs the edges in the form of a list, where each element is a dictionary that meets the requirements of the element schema. At the same time, the large language model can be prompted to generate supplementary edge extraction results. In generation, the model relies only on the identified nodes and infers the relationships between nodes through intrinsic knowledge, even if these relationships are not explicitly expressed in the text. This approach makes full use of the pre-trained knowledge and reasoning capabilities of the large language model.

[0100] It should be noted that there is a large degree of randomness in the generation process of the large language model. Even if the temperature parameter is set to 0, the large language model may still give different results in multiple generations. In order to obtain a more reliable extraction process, an exception handling method can be designed, mainly for the situation where the output format of the large language model does not conform to the element schema and cannot be parsed. In this case, the large language model often extracts the main information, but the format is inconsistent with the defined node or edge attributes. By combining the output containing the problem and the format requirements into another prompt template, the large language model is required to modify the output according to the schema and set the maximum number of retries. Through the exception handling method, the parsing problems of the output format can be significantly reduced.

[0101] According to the foregoing description, the decision-making ability evaluation method provided by the present invention obtains a group discussion text, wherein the group discussion text is formed according to the participation discussion of the participants to be evaluated; based on the group discussion text, a participant influence map corresponding to the participants to be evaluated is obtained, wherein the participant influence map includes map nodes and map edges corresponding to the behaviors formed by the participants to be evaluated in the process of participating in the group discussion; based on the participant influence map, the decision-making ability of the participants to be evaluated under the preset ability evaluation indicators is evaluated. The decision-making ability that can automatically and objectively evaluate the individual ability in the group discussion is realized, which is of great significance for improving the quality of group discussion and optimizing the decision-making process.

[0102] The decision-making ability assessment device provided by the present invention is described below. The decision-making ability assessment device described below and the decision-making ability assessment method described above can be referenced to each other.

[0103] Figure 6It is a structural schematic diagram of the decision-making ability evaluation device provided by the present invention.

[0104] The following will be combined Figure 6 The structure of the decision-making ability evaluation device provided by the present invention is described.

[0105] In an exemplary embodiment of the present invention, Figure 6 It can be seen that the decision-making ability assessment device may include an acquisition module 610, a processing module 620, and an evaluation module 630, and each module will be introduced below.

[0106] The acquisition module 610 may be configured to acquire a group discussion text, wherein the group discussion text is formed according to the discussion participated in by the participants to be evaluated; The processing module 620 may be configured to obtain a participant influence map corresponding to the participant to be evaluated based on the group discussion text, wherein the participant influence map includes map nodes and map edges corresponding to the behaviors formed by the participant to be evaluated during the group discussion; The evaluation module 630 may be configured to evaluate the decision-making ability of the participant to be evaluated under preset ability evaluation indicators based on the participant influence map.

[0107] In an exemplary embodiment of the present invention, when the participant influence map includes a preset sub-map structure, the evaluation module 630 can implement the ability evaluation of the decision-making ability of the participant to be evaluated under the preset ability evaluation index based on the participant influence map in the following manner: Based on the participant influence map, extracting a preset subgraph structure in the participant influence map; Calling a preset standard subgraph structure, wherein the standard subgraph structure is a subgraph structure corresponding to a participant having decision-making ability under a preset ability evaluation index; Based on the preset subgraph structure and the standard subgraph structure, the decision-making ability of the participant to be evaluated is evaluated under the preset ability evaluation index.

[0108] In an exemplary embodiment of the present invention, the evaluation module 630 may implement the ability evaluation of the decision-making ability of the participant to be evaluated under the preset ability evaluation index based on the preset subgraph structure and the standard subgraph structure in the following manner: In the case where the preset subgraph structure matches the standard subgraph structure, determining that the participant to be evaluated has decision-making ability under the preset ability evaluation index; In the case that the preset subgraph structure does not match the standard subgraph structure, it is determined that the participant to be evaluated does not have the decision-making ability under the preset ability evaluation index.

[0109] In an exemplary embodiment of the present invention, when the participant influence map does not include a preset sub-graph structure, the evaluation module 630 may implement the following method to evaluate the decision-making ability of the participant to be evaluated under the preset ability evaluation index based on the participant influence map: Constructing a first prompt text, wherein the first prompt text is used to represent target node information for extracting decision-making ability under a preset ability evaluation indicator from a participant influence map; Call the generative pre-trained transformer model; Inputting the participant influence map and the first prompt text into the generative pre-trained transformer model to obtain the target node information output by the generative pre-trained transformer model; Based on the target node information, the decision-making ability of the participant to be evaluated is evaluated under the preset ability evaluation index.

[0110] In an exemplary embodiment of the present invention, the evaluation module 630 can implement the ability evaluation of the decision-making ability of the participant to be evaluated under the preset ability evaluation index based on the target node information in the following manner: In the case where the target node information meets the preset node information requirements, determining that the participant to be evaluated has decision-making ability under the preset ability evaluation index; In the case where the target node information does not meet the preset node information requirements, it is determined that the participant to be evaluated does not have the decision-making ability under the preset ability evaluation index, wherein: The preset node information requirements are node information requirements corresponding to the decision-making capabilities of the participant to be evaluated under the preset capability evaluation indicators.

[0111] In an exemplary embodiment of the present invention, when the participant influence map does not include a preset sub-graph structure, the evaluation module 630 may implement the following method to evaluate the decision-making ability of the participant to be evaluated under the preset ability evaluation index based on the participant influence map: Based on the participant influence map, determine the target node information in the participant influence map, wherein the target node information is used to represent the node information in the participant influence map used to evaluate the decision-making ability under the preset ability evaluation index; In the case where the target node information meets the preset node information requirements, determining that the participant to be evaluated has decision-making ability under the preset ability evaluation index; In the case where the target node information does not meet the preset node information requirements, it is determined that the participant to be evaluated does not have the decision-making ability under the preset ability evaluation index, wherein: The preset node information requirements are node information requirements corresponding to the decision-making capabilities of the participant to be evaluated under the preset capability evaluation indicators.

[0112] In an exemplary embodiment of the present invention, the processing module 620 may obtain a participant influence map corresponding to the participant to be evaluated based on the group discussion text in the following manner: Based on the group discussion text, respectively obtaining graph nodes with a preset schema structure corresponding to the group discussion text and graph edges with a preset schema structure corresponding to the group discussion text; Based on the graph nodes with a preset schema structure and the graph edges with a preset schema structure, a participant influence graph corresponding to the participant to be evaluated is obtained.

[0113] In an exemplary embodiment of the present invention, the processing module 620 may obtain a graph node with a preset graph structure corresponding to the group discussion text based on the group discussion text in the following manner: Constructing a second prompt text, wherein the second prompt text is used to represent and output a graph node with a preset graph structure corresponding to the group discussion text, wherein the graph node at least includes a chance node, a decision node, and a utility node; Call the generative pre-trained transformer model; The group discussion text and the second prompt text are input into the generative pre-trained transformer model to obtain a graph node with a preset schema structure corresponding to the group discussion text output by the generative pre-trained transformer model.

[0114] In an exemplary embodiment of the present invention, the processing module 620 may obtain a graph edge with a preset schema structure corresponding to the group discussion text based on the group discussion text in the following manner: Constructing a third prompt text, wherein the third prompt text is used to represent and output a graph edge with a preset schema structure corresponding to the group discussion text; Call the generative pre-trained transformer model; The group discussion text, the graph node with a preset schematic structure, and the third prompt text are input into the generative pre-trained transformer model to obtain the graph edge with a preset schematic structure corresponding to the group discussion text output by the generative pre-trained transformer model.

[0115] Figure 7 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 7As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730 and a communication bus 740, wherein the processor 710, the communication interface 720 and the memory 730 communicate with each other through the communication bus 740. The processor 710 may call the logic instructions in the memory 730 to execute the decision-making ability evaluation method, which includes: obtaining a group discussion text, wherein the group discussion text is formed according to the participation discussion of the participants to be evaluated; based on the group discussion text, obtaining a participant influence map corresponding to the participants to be evaluated, wherein the participant influence map includes map nodes and map edges corresponding to the behaviors formed by the participants to be evaluated in the process of participating in the group discussion; based on the participant influence map, the decision-making ability of the participants to be evaluated under the preset ability evaluation index is evaluated.

[0116] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0117] On the other hand, the present invention also provides a computer program product, which includes a computer program, and the computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the decision-making ability assessment method provided by the above-mentioned methods, and the method includes: obtaining a group discussion text, wherein the group discussion text is formed according to the participation discussion of the participants to be evaluated; based on the group discussion text, obtaining a participant influence map corresponding to the participants to be evaluated, wherein the participant influence map includes map nodes and map edges corresponding to the behaviors formed by the participants to be evaluated in the process of participating in the group discussion; based on the participant influence map, the decision-making ability of the participants to be evaluated under preset ability evaluation indicators is evaluated.

[0118] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the decision-making ability assessment method provided by the above-mentioned methods, the method comprising: obtaining a group discussion text, wherein the group discussion text is formed according to the participation discussion of the participants to be evaluated; based on the group discussion text, obtaining a participant influence map corresponding to the participants to be evaluated, wherein the participant influence map includes map nodes and map edges corresponding to the behaviors formed by the participants to be evaluated in the process of participating in the group discussion; based on the participant influence map, conducting an ability evaluation on the decision-making ability of the participants to be evaluated under preset ability evaluation indicators.

[0119] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0120] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0121] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A decision-making ability assessment method, characterized in that: The method comprises: Acquiring a group discussion text, wherein the group discussion text is formed according to the participation and discussion of the participants to be evaluated; Based on the group discussion text, a participant influence map corresponding to the participant to be evaluated is obtained, wherein the participant influence map includes map nodes and map edges corresponding to the behaviors formed by the participant to be evaluated during the group discussion; Based on the participant influence map, the decision-making ability of the participant to be evaluated is evaluated under the preset ability evaluation indicators.

2. The decision-making ability assessment method according to claim 1, characterized in that: In the case where the participant influence map includes a preset sub-map structure, the decision-making ability of the participant to be evaluated under the preset ability evaluation index is evaluated based on the participant influence map, specifically including: Based on the participant influence map, extracting a preset subgraph structure in the participant influence map; Calling a preset standard subgraph structure, wherein the standard subgraph structure is a subgraph structure corresponding to a participant having decision-making ability under a preset ability evaluation index; Based on the preset subgraph structure and the standard subgraph structure, the decision-making ability of the participant to be evaluated is evaluated under the preset ability evaluation index.

3. The decision-making ability assessment method according to claim 2, characterized in that: The ability evaluation of the decision-making ability of the participant to be evaluated under the preset ability evaluation index based on the preset subgraph structure and the standard subgraph structure specifically includes: In the case where the preset subgraph structure matches the standard subgraph structure, determining that the participant to be evaluated has decision-making ability under the preset ability evaluation index; In the case that the preset subgraph structure does not match the standard subgraph structure, it is determined that the participant to be evaluated does not have the decision-making ability under the preset ability evaluation index.

4. The decision-making ability assessment method according to claim 1, characterized in that: In the case where the participant influence map does not include a preset sub-graph structure, before evaluating the decision-making ability of the participant to be evaluated under the preset ability evaluation index based on the participant influence map, the method further includes: Constructing a first prompt text, wherein the first prompt text is used to represent target node information for extracting decision-making ability under a preset ability evaluation indicator from a participant influence map; The decision-making ability of the participant to be evaluated under the preset ability evaluation index is evaluated based on the participant influence map, specifically including: Call the generative pre-trained transformer model; Inputting the participant influence map and the first prompt text into the generative pre-trained transformer model to obtain the target node information output by the generative pre-trained transformer model; Based on the target node information, the decision-making ability of the participant to be evaluated is evaluated under the preset ability evaluation index.

5. The decision-making ability assessment method according to claim 4, characterized in that: The step of evaluating the decision-making ability of the participant to be evaluated based on the target node information under the preset ability evaluation index specifically includes: In the case where the target node information meets the preset node information requirements, determining that the participant to be evaluated has decision-making ability under the preset ability evaluation index; In the case where the target node information does not meet the preset node information requirements, it is determined that the participant to be evaluated does not have the decision-making ability under the preset ability evaluation index, wherein: The preset node information requirements are node information requirements corresponding to the decision-making capabilities of the participant to be evaluated under the preset capability evaluation indicators.

6. The decision-making ability assessment method according to claim 1, characterized in that: In the case where the participant influence map does not include a preset sub-map structure, the decision-making ability of the participant to be evaluated under the preset ability evaluation index is evaluated based on the participant influence map, specifically including: Based on the participant influence map, determine the target node information in the participant influence map, wherein the target node information is used to represent the node information in the participant influence map used to evaluate the decision-making ability under the preset ability evaluation index; In the case where the target node information meets the preset node information requirements, determining that the participant to be evaluated has decision-making ability under the preset ability evaluation index; In the case where the target node information does not meet the preset node information requirements, it is determined that the participant to be evaluated does not have the decision-making ability under the preset ability evaluation index, wherein: The preset node information requirements are node information requirements corresponding to the decision-making capabilities of the participant to be evaluated under the preset capability evaluation indicators.

7. The decision-making ability assessment method according to any one of claims 1 to 6, characterized in that: The obtaining, based on the group discussion text, a participant influence map corresponding to the participant to be evaluated specifically includes: Based on the group discussion text, respectively obtaining graph nodes with a preset schema structure corresponding to the group discussion text and graph edges with a preset schema structure corresponding to the group discussion text; Based on the graph nodes with a preset schema structure and the graph edges with a preset schema structure, a participant influence graph corresponding to the participant to be evaluated is obtained.

8. The decision-making ability assessment method according to claim 7, characterized in that: The obtaining, based on the group discussion text, a graph node having a preset graph structure corresponding to the group discussion text specifically includes: Constructing a second prompt text, wherein the second prompt text is used to represent and output a graph node with a preset graph structure corresponding to the group discussion text, wherein the graph node at least includes a chance node, a decision node, and a utility node; Call the generative pre-trained transformer model; The group discussion text and the second prompt text are input into the generative pre-trained transformer model to obtain a graph node with a preset schema structure corresponding to the group discussion text output by the generative pre-trained transformer model.

9. The decision-making ability assessment method according to claim 8, characterized in that: The obtaining, based on the group discussion text, a graph edge having a preset schema structure corresponding to the group discussion text specifically includes: Constructing a third prompt text, wherein the third prompt text is used to represent and output a graph edge with a preset schema structure corresponding to the group discussion text; Call the generative pre-trained transformer model; The group discussion text, the graph node with a preset schematic structure, and the third prompt text are input into the generative pre-trained transformer model to obtain the graph edge with a preset schematic structure corresponding to the group discussion text output by the generative pre-trained transformer model.

10. A decision-making ability assessment device, characterized in that: The device comprises: An acquisition module, used for acquiring a group discussion text, wherein the group discussion text is formed according to the participation and discussion of the participants to be evaluated; A processing module, configured to obtain a participant influence map corresponding to the participant to be evaluated based on the group discussion text, wherein the participant influence map includes map nodes and map edges corresponding to the behaviors formed by the participant to be evaluated during the group discussion; An evaluation module is used to evaluate the decision-making ability of the participant to be evaluated under preset ability evaluation indicators based on the participant influence map.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the decision-making ability assessment method according to any one of claims 1 to 9 is implemented.

12. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the decision-making ability assessment method according to any one of claims 1 to 9 is implemented.

13. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the decision-making ability assessment method according to any one of claims 1 to 9 is implemented.

Images