Electronic brainstorming support system based on creativity scoring mechanism
By integrating a creativity scoring mechanism and inspiration cards into an electronic brainstorming system, and utilizing SSD and CNN networks to identify facial emotions, dynamically score, and distribute inspiration cards, the problem of insufficient group emotion recognition in electronic brainstorming is solved, thereby improving team creativity and innovation efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- ZHEJIANG UNIV
- Filing Date
- 2022-12-08
- Publication Date
- 2026-06-09
AI Technical Summary
Existing electronic brainstorming systems lack convenient methods for recognizing group emotions, resulting in insufficient support for creativity and impacting team innovation efficiency.
An electronic brainstorming support system based on a creativity scoring mechanism was designed. Through a modular architecture of server and client, it uses SSD and CNN networks to identify individual facial emotions, combines group emotions and solution update frequency, dynamically scores, and randomly distributes inspiration cards related to the topic to stimulate creativity.
Effective identification and intervention of team emotions improved group creativity, enhanced the innovative effect of brainstorming, stimulated creativity through inspiration cards, and improved the team's collaborative innovation capabilities.
Smart Images

Figure CN116186448B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a brainstorming support system, specifically an electronic brainstorming support system based on a creativity scoring mechanism. Background Technology
[0002] Brainstorming is a common creative stimulation method that aims to generate numerous ideas that may solve a given problem, relying on the creativity of participants. In 1992, Gallupe et al. designed an electronic brainstorming system (EBS) that allowed group members to interact and communicate via computers, thereby improving collaborative innovation capabilities. In EBS, group members communicated through their respective computers; a dialog box at the top of the screen displayed the ideas of other group members, while a dialog box at the bottom allowed members to input their own ideas. This overcame the "productivity blockage" and evaluation anxieties that inevitably occur in face-to-face brainstorming, effectively improving team performance. Despite the many benefits of electronic brainstorming and its suitability for the era of globalized office collaboration, there is currently no convenient and universally applicable electronic brainstorming support system. People typically use ordinary meeting software or chat-like electronic brainstorming platforms for online brainstorming activities.
[0003] On the other hand, the key to electronic brainstorming remains supporting the participants' creativity. Among the many variables proven to predict creativity, emotion has been verified as one of the least controversial predictors (George & Brief, 1996; Isen & Baron, 1991; Mumford, 2003). Creativity researchers believe that positive emotions increase creative performance, while negative emotions decrease creativity. According to research by social psychologists in the field of psychology, collective emotions arise from and influence the emotional states of individuals within a group. Currently, common emotion recognition methods are mainly divided into two categories: recognition based on non-physiological signals such as facial expressions and tone of voice, and recognition based on physiological signals such as heart rate data or brainwave data. Current solutions include emotion recognition in one-on-one video conferences, such as Chinese patent document CN107636684A, which uses template matching for facial localization and emotion classification; others use wearable devices to transmit digital signals, which are then processed and analyzed to obtain emotion classification, as described in Chinese patent document CN108932951A; and still others use traditional methods such as text feedback. However, these methods all have limitations, being inconvenient and inaccurate. Therefore, how to design an effective group emotion determination method for electronic brainstorming is the main technical problem that this invention aims to solve. Summary of the Invention
[0004] In view of the problems existing in the prior art, the purpose of this invention is to provide an electronic brainstorming support system based on a creativity scoring mechanism.
[0005] This invention is specifically implemented through the following technical solutions:
[0006] An electronic brainstorming support system based on a creativity scoring mechanism, which includes a server and a client;
[0007] The server-side component consists of a web module, a creativity scoring module, and an inspiration card module. The creativity scoring module comprises an emotion scoring module and a solution pool scoring module. The web module is primarily responsible for receiving and responding to HTTP requests from clients. The emotion scoring module mainly uses a dynamic emotion scoring mechanism to analyze and score collective emotions.
[0008] The client consists of a client web module, an emotion recognition module, and a solution pool module. The client web module is mainly responsible for sending the user ID and corresponding emotion level to the server via HTTP. The emotion recognition module is mainly responsible for extracting facial regions and classifying emotions from images.
[0009] Furthermore, after the client collects image information and classifies emotions, it scores the collective emotion level through the mapping relationship between individual emotions and collective emotions on the communication network. At the same time, the client uploads brainstorming results to the solution pool. The communication network records the solution update frequency and scores the update frequency according to preset standards. When the collective emotion score and creativity level score drop to a certain standard, the system transmits the scoring results to the server (EBS) and drives the server to randomly distribute "inspiration cards" to users every two minutes according to the brainstorming topic.
[0010] Furthermore, individual emotions are acquired by using SSD to locate the face, capturing facial images, and then classifying emotions using a CNN network. The classification method is as follows: individual emotions are classified into three categories with a total of eight emotions, including positive emotions, negative emotions, and neutral emotions. Positive emotions include happiness and relaxation, negative emotions include anxiety, anger, boredom, fear, and sadness, and neutral emotions include boredom and surprise.
[0011] Furthermore, individual emotions are described by the percentage of each emotion, and the formula for describing an individual's dynamic emotions is:
[0012] P={tendency, happiness, relaxation, anxiety, anger, fear, sadness, boredom, surprise}={t1, h1, r1, anx1, ang1, f1, s1, b1, su1};
[0013] P represents the set of individual emotion descriptions, where tendency represents emotional tendency, happiness represents happiness, relaxation represents relaxation, anxiety represents anxiety, ange represents anger, fear represents fear, sadness represents sadness, boredom represents boredom, and surprise represents surprise. t1 represents user 1's emotional tendency, h1 represents the percentage of user 1's happiness, r1 represents the percentage of user 1's happiness, anx1 represents the percentage of user 1's relaxation, ang1 represents the percentage of user 1's anxiety, f1 represents the percentage of user 1's fear, s1 represents the percentage of user 1's sadness, b1 represents the percentage of user 1's boredom, and su1 represents the percentage of user 1's surprise.
[0014] Furthermore, the dynamic score of collective emotion is: collective emotion is the sum of the proportions of each emotion among all encrypted individual emotions, with a cycle of 3 seconds. The formula describing collective emotion is:
[0015] G={g-happiness, g-relaxation, g-anxiety, g-anger, g-fear, g-sadness, g-boredom, g-surprise}.
[0016] Furthermore, based on the current collective creativity, the electronic brainstorming team members are randomly assigned inspiration cards that can spark inspiration. Specifically, if the sum of negative emotions in the collective emotional creativity is greater than 50% or the update frequency of the solution pool is less than one solution produced per person every three minutes on average, it is defined as "low group creativity". The formula for describing group creativity is: C={{group emotion},{frequency}}, which describes the current group emotion and solution update frequency of the brainstorming group.
[0017] Furthermore, inspiration cards are informational cue cards that align with the current brainstorming theme. These cards feature simple pictures and short phrases, allowing team members to expand their associations and further develop their creative ideas.
[0018] This invention is based on the basic framework of electronic brainstorming and combines inspiration cards driven by group emotions and creativity to form a new computer collaborative innovation support tool—the electronic brainstorming support system. This system randomly outputs inspiration cards to electronic brainstorming team members based on the current group creative ability (group emotions and solution pool update frequency), and further utilizes inspiration cards as a creative support tool to stimulate group creative ability.
[0019] This invention can identify the collective emotions of members of an electronic brainstorming team and, in conjunction with group creative abilities, provide positive intervention for the current brainstorming session. Attached Figure Description
[0020] Figure 1 This is a system architecture diagram of the present invention;
[0021] Figure 2 Flowchart of the dynamic emotion scoring mechanism;
[0022] Figure 3 A structural diagram of the server and client;
[0023] Figure 4 Here is a flowchart of the emotion recognition module's workflow;
[0024] Figure 5 This is a diagram of an inspiration card;
[0025] Figure 6 This is a flowchart illustrating the implementation steps of the present invention. Detailed Implementation
[0026] The present invention will be further described in detail below with reference to the accompanying drawings to provide a better understanding of the technical solution.
[0027] System Architecture: The system is built using a web service. Clients (team members) collect image information and categorize emotions. (To ensure data accuracy), three minutes after the brainstorming session begins, a collective emotion level score is assigned via a communication network based on the mapping relationship between individual and collective emotions. Simultaneously, clients upload brainstorming results to a solution pool. The communication network records the solution update frequency and scores the update frequency according to preset criteria. When the collective emotion score and creativity level score drop to a certain standard, the system transmits the scoring results to the server (EBS) and drives the server to randomly distribute "inspiration cards" to users every two minutes based on the brainstorming topic.
[0028] Individual Emotion Acquisition: Facial images are captured using SSD localization and then classified using a CNN network. The classification method categorizes individual emotions into three groups, totaling eight emotions. Specifically, these are: positive emotions (happiness and relaxation); negative emotions (anxiety, anger, boredom, fear, and sadness); and neutral emotions (boredom and surprise). These emotions have been shown to be common and can negatively impact learning efficiency. In practical application, after brainstorming begins (3 minutes later), the emotion detection device continuously collects image information from participants. The collected images remain on the client-side camera. Emotion recognition is performed on the client side. Upon receiving the data, the server updates the emotion level dictionary and the group emotion level, and combines this with the update frequency of the solution pool to score creativity. Based on the score, "inspiration cards" are distributed to users.
[0029] Individual emotions are described by the percentage of each emotion. The formula for describing an individual's dynamic emotions is:
[0030] P={tendency, happiness, relaxation, anxiety, anger, fear, sadness, boredom, surprise}={t1, h1, r1, anx1, ang1, f1, s1, b1, su1}
[0031] P represents the set of individual emotion descriptions, where tendency represents emotion tendency, happiness represents happiness, relaxation represents relaxation, anxiety represents anxiety, ange represents anger, fear represents fear, sadness represents sadness, boredom represents boredom, and surprise represents surprise. t1 represents user 1's emotion tendency, h1 represents the percentage of user 1's happiness, r1 represents the percentage of user 1's happiness, anx1 represents the percentage of user 1's relaxation, ang1 represents the percentage of user 1's anxiety, f1 represents the percentage of user 1's fear, s1 represents the percentage of user 1's sadness, b1 represents the percentage of user 1's boredom, and su1 represents the percentage of user 1's surprise.
[0032] Group dynamic sentiment scoring: Group sentiment is the sum of the proportions of each sentiment among all encrypted individual sentiments, with a cycle of 3 seconds. The formula for describing group sentiment is:
[0033] G={g-happiness,g-relaxation, g-anxiety,g-anger,g-fear,g-sadness, g-boredom,g-surprise};
[0034] Group sentiment is the sum of the proportions of each individual sentiment in all encrypted personal sentiments. For example, g-happiness represents the sum of happy sentiments of each user.
[0035] Group Creativity Scoring: This invention randomly generates "inspiration cards" to inspire team members in electronic brainstorming sessions based on the current group's creative capacity (group sentiment and solution pool update frequency). Specifically, a "low group creativity" is defined as one where the sum of negative emotions in the group's emotional creativity exceeds 50% or the solution pool update frequency is lower than an average of one solution output per person every three minutes. The formula for describing group creativity is: C = {{group sentiment}, {frequency}}, which describes the current group sentiment and solution update frequency of the brainstorming group. When group creativity is low, the system transmits the creativity score to the server (EBS). The server randomly distributes "inspiration cards" to users every two minutes based on the brainstorming topic, thereby stimulating the group to engage in a new round of brainstorming. After a series of iterative designs and optimizations, this invention merges the elements into a concise list of inspiration cards. An "inspiration card" is an informational cue card consistent with the current brainstorming topic, containing simple images and short words. Team members can use the images and text to expand their associations and further generate creative ideas.
[0036] The computer-based collaborative innovation support tool proposed in this invention—the Electronic Brainstorming Support System—can help electronic brainstorming teams generate ideas and solutions more efficiently in various fields and scenarios.
[0037] The system framework diagram of this invention is as follows: Figure 1 As shown, each client is connected to an external camera. The image data captured by the camera is transmitted to the client, which then performs image classification on the video source to obtain the emotion classification for each user. Simultaneously, the client sends the generated emotion classification and the corresponding user ID to the server via HTTP protocol. After receiving the data, the server updates the emotion level dictionary and the group emotion level, and scores the group's creativity based on the update frequency of the solution pool. Based on the scoring results, "inspiration cards" are distributed to users.
[0038] Its workflow diagram is as follows Figure 2 As shown, at the start of the meeting, the dictionary is initialized to an empty dictionary. Then, the client sends the user ID and its corresponding emotion category to the server, and the dictionary stores this data. If the ID does not exist in the dictionary, it is used as the key and the corresponding emotion category as the value. If the ID exists, the emotion category corresponding to that ID is updated to the newly passed emotion category.
[0039] The server and client structures are as follows: Figure 3As shown, the server consists of three modules: a Web module, a creativity scoring module, and an inspiration card module. The creativity scoring module comprises an emotion scoring module and a solution pool scoring module. The Web module is primarily responsible for receiving and responding to HTTP requests from clients. The emotion scoring module mainly uses a dynamic emotion scoring mechanism to analyze and score collective emotions.
[0040] The client consists of three modules: a web module, an emotion recognition module, and a solution pool module. The web module is primarily responsible for sending the user ID and corresponding emotion level to the server via HTTP. The emotion recognition module mainly performs facial region extraction and emotion classification on the image.
[0041] The workflow of the emotion recognition module is as follows: Figure 4 As shown, for image data captured by the camera, this module first calls the YOLO object detection network with an improved attention mechanism to extract facial regions, and then uses the GoogleNet network to classify facial emotions to obtain the emotion classification corresponding to the image.
[0042] Electronic Brainstorming Support System Inspiration Cards: After user research and a series of iterative designs and optimizations, this invention merges elements into a concise inspiration card design space. An "inspiration card" is an informational cue card consistent with the current brainstorming topic, featuring simple images and brief words. Team members can use the images and text to expand their associations and further generate creative ideas. The inspiration card system can be built in stages and continuously enriched and expanded based on the thinking processes of users of the electronic brainstorming support system, ultimately perfecting our brainstorming support system through participatory design. To standardize and document the inspiration cards, this invention constructs an inspiration card design space. Taking the furniture sector within the GICS industry as an example, the inspiration cards randomly provide clues to brainstorming members from three dimensions: materials, function, and structure, such as... Figure 5 As shown.
[0043] The steps for implementing this invention are as follows: Figure 6 As shown, specifically:
[0044] 1) The system is built using web services. The client (team members) uses a camera to collect image data. The face extraction module uses the YOLO target detection network and attention mechanism to analyze facial information and classifies emotions based on the GoogleNet classification network.
[0045] 1.1) Individual Emotion Acquisition Method: Facial images were captured using SSD localization and then classified using a CNN network. Specifically, individual emotions were categorized into three classes with a total of eight emotions. Positive emotions included happiness and relaxation; negative emotions included anxiety, anger, boredom, fear, and sadness; and neutral emotions included boredom and surprise. These emotions have been shown to be relatively common and can affect learning efficiency.
[0046] 1.2) Group dynamic emotion scoring: Group emotion is the sum of the proportions of each emotion in all encrypted individual emotions, with a cycle of 3 seconds;
[0047] 1.3) Dynamic Emotion Scoring Mechanism: At the start of the meeting, the dictionary is initialized to an empty dictionary. Then, the client sends the user ID and its corresponding emotion category to the server, and the dictionary stores this data. If the ID does not exist in the dictionary, it is stored as the key and the corresponding emotion category as the value. If the ID exists, the emotion category corresponding to that ID is updated to the newly passed emotion category.
[0048] 2) Calculate the total proportion of different emotions in the group to obtain the group emotion score: In order to ensure the authenticity and reliability of the data, 3 minutes after the brainstorming begins, the group emotion level is dynamically scored in the web module through the mapping relationship between individual emotions and collective emotions.
[0049] 3) Build a system using web services, accumulate brainstormed solutions in a solution pool, and analyze the solution update frequency;
[0050] 4) Group Creativity Rating: The formula for describing group creativity is: C = {{group emotion}, {frequency}}, which describes the group's emotional state and the frequency of solution updates. Specifically, if the sum of negative emotions in group emotional creativity is greater than 50% or the solution pool update frequency is less than one solution produced per person every three minutes on average, it is defined as "low group creativity". Conversely, it is defined as "non-low group creativity".
[0051] 5) When group creativity is low, the system transmits the creativity score results to the server (EBS). The server randomly distributes "inspiration cards" to users every two minutes based on the brainstorming topic, thereby stimulating the group to conduct a new round of brainstorming. The "inspiration card" is an information prompt card consistent with the current brainstorming topic. The card has simple pictures and short words, and team members can expand their associations based on the pictures and text to further generate creative ideas;
[0052] 6) The inspiration card system can be built in stages. In addition to the design concept of this invention, it can be continuously enriched and expanded based on people's ideas when using the electronic brainstorming support system, and finally perfected in the form of participatory design.
Claims
1. An electronic brainstorming support system based on a creativity scoring mechanism, comprising a server and a client, characterized by: The server-side component consists of a server-side web module, a creativity scoring module, and an information prompt card module. The creativity scoring module comprises an emotion scoring module and a solution pool scoring module. The server-side web module is primarily responsible for receiving and responding to HTTP requests from clients. The emotion scoring module mainly uses a dynamic emotion scoring mechanism to analyze and score collective emotions. The client consists of a web module, an emotion recognition module, and a solution pool module. The web module is mainly responsible for sending the user ID and corresponding emotion level to the server via HTTP. The emotion recognition module is mainly responsible for extracting facial regions and classifying emotions from images. After collecting image information and classifying emotions, the client scores the collective emotion level through the mapping relationship between individual emotions and collective emotions on the communication network. At the same time, the client uploads brainstorming results to the solution pool. The communication network records the solution update frequency and scores the update frequency according to preset standards. When the collective emotion score and creativity level score drop to a certain standard, the system transmits the scoring results to the server (EBS) and drives the server to randomly distribute "information prompt cards" to users every two minutes according to the brainstorming topic.
2. The electronic brainstorming support system based on a creativity scoring mechanism as described in claim 1, characterized in that... Individual emotions are captured by localizing the face using SSD, extracting facial images, and then classifying emotions using a CNN network. The classification method categorizes individual emotions into three categories with a total of eight emotions: positive emotions, negative emotions, and neutral emotions. Positive emotions include happiness and relaxation, negative emotions include anxiety, anger, boredom, fear, and sadness, and neutral emotions include boredom and surprise.
3. The electronic brainstorming support system based on a creativity scoring mechanism as described in claim 2, characterized in that... Individual emotions are described by the percentage of each emotion. The formula for describing an individual's dynamic emotions is: P={tendency, happiness, relaxation, anxiety, anger, fear, sadness, boredom, surprise}={t1, h1, r1, anx1, ang1, f1, s1, b1, su1}; P represents the set of individual emotion descriptions, tendency represents emotion tendency, happiness represents happiness, relaxation represents relaxation, anxiety represents anxiety, ange represents anger, fear represents fear, sadness represents sadness, boredom represents boredom, and surprise represents surprise; t1 represents user 1's emotion tendency, h1 represents the percentage of user 1's happiness, r1 represents the percentage of user 1's happiness, anx1 represents the percentage of user 1's relaxation, ang1 represents the percentage of user 1's anxiety, f1 represents the percentage of user 1's fear, s1 represents the percentage of user 1's sadness, b1 represents the percentage of user 1's boredom, and su1 represents the percentage of user 1's surprise.
4. The electronic brainstorming support system based on a creativity scoring mechanism as described in claim 1, characterized in that... The dynamic score of collective emotion is: collective emotion is the sum of the proportions of each emotion among all encrypted individual emotions, with a cycle of 3 seconds. The formula describing collective emotion is: G={g-happiness, g-relaxation, g-anxiety, g-anger, g-fear, g-sadness, g-boredom, g-surprise}.
5. The electronic brainstorming support system based on a creativity scoring mechanism as described in claim 1, characterized in that... Based on the current collective creativity, the electronic brainstorming team members are randomly given information prompt cards that can inspire creativity. Specifically, if the sum of negative emotions in the collective emotional creativity is greater than 50% or the update frequency of the solution pool is less than one solution produced per person every three minutes, it is defined as "low group creativity". The formula for describing group creativity is: C = {{groupemotuon},{frequency}}, which describes the current group emotion and solution update frequency of the brainstorming group.
6. The electronic brainstorming support system based on a creativity scoring mechanism as described in claim 1, characterized in that... Information cue cards are information cue cards that are consistent with the current brainstorming theme. The cards have simple pictures and short words, which can help team members expand their associations and further generate creative ideas.