Recently, with the advancement of technologies in AI and Knowledge Base, several museums are using chatbots for visitors. One of the problems with these technologies, however is that gradually tends to be of no real interest to visitors owing to the lack of significant interaction, this eventually distracts visitors from experiencing the exhibits. In the demo, we present AIMuBot, an interactive system for searching the information from the museum's knowledge base with natural language. The system has the following characteristics: (1) It supports natural language voice-based interaction with the visitors to ask questions; (2) It provides a voice-based graphical interface to help visitors refine the questions. (3) It retrieves information from the knowledge base for the visitors.
{"title":"An AI chatbot for the museum based on user Interaction over a knowledge base","authors":"Chunyan Zhou, Baivab Sinha, Minghua Liu","doi":"10.1145/3421766.3421888","DOIUrl":"https://doi.org/10.1145/3421766.3421888","url":null,"abstract":"Recently, with the advancement of technologies in AI and Knowledge Base, several museums are using chatbots for visitors. One of the problems with these technologies, however is that gradually tends to be of no real interest to visitors owing to the lack of significant interaction, this eventually distracts visitors from experiencing the exhibits. In the demo, we present AIMuBot, an interactive system for searching the information from the museum's knowledge base with natural language. The system has the following characteristics: (1) It supports natural language voice-based interaction with the visitors to ask questions; (2) It provides a voice-based graphical interface to help visitors refine the questions. (3) It retrieves information from the knowledge base for the visitors.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128944833","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Flexible learning is based on studying learners' psychology and behavior, using the principle of flexible management and flexible decision-making and adopting the mode of flexible contingency thinking to transform the will of teaching organizers into learners' conscious behavior through imperceptible change and cultivation. Flexible learning model can solve the problems, such as short of innovation, weak autonomy and lack of collective learning atmosphere, existing in network education. This paper introduces the concept and mechanism of flexible learning in the SPOC (Small Private Online Course), relies on big data and intelligent technology, optimizes the design of SPOC platform architecture, and arranges a series of interactive learning sites with the community interface of Knowledge Farm, and finally, constructs the learning situation of a multi-level, three-dimensional intelligent SPOC game.
{"title":"Scenario Design of Intelligent SPOC Knowledge Farm Based on Flexible Learning","authors":"Huangxing Zeng","doi":"10.1145/3421766.3421784","DOIUrl":"https://doi.org/10.1145/3421766.3421784","url":null,"abstract":"Flexible learning is based on studying learners' psychology and behavior, using the principle of flexible management and flexible decision-making and adopting the mode of flexible contingency thinking to transform the will of teaching organizers into learners' conscious behavior through imperceptible change and cultivation. Flexible learning model can solve the problems, such as short of innovation, weak autonomy and lack of collective learning atmosphere, existing in network education. This paper introduces the concept and mechanism of flexible learning in the SPOC (Small Private Online Course), relies on big data and intelligent technology, optimizes the design of SPOC platform architecture, and arranges a series of interactive learning sites with the community interface of Knowledge Farm, and finally, constructs the learning situation of a multi-level, three-dimensional intelligent SPOC game.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127894426","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mathematical model of ship steam turbine unit including steam turbine, gear reducer and condenser is established. Given the reasonable boundary conditions, the weight and volume multi-objective optimization of a typical ship steam turbine is carried out by using a modified optimization algorithm to harmonize the selected design variables. The results show that the weight and volume of the optimized ship steam turbine unit decrease by 4.2% and 6.3% respectively, which demonstrates the capability of the optimization method in optimizing the weight and volume of ship steam turbine unit.
{"title":"Multi-Objective Optimization of Ship Steam Turbine Unit","authors":"Cheng Wang, Zhengming Tang, Xiang Wan, K. Cheng, Haijun Sun, Yuan Fang, Shan Gao","doi":"10.1145/3421766.3421893","DOIUrl":"https://doi.org/10.1145/3421766.3421893","url":null,"abstract":"Mathematical model of ship steam turbine unit including steam turbine, gear reducer and condenser is established. Given the reasonable boundary conditions, the weight and volume multi-objective optimization of a typical ship steam turbine is carried out by using a modified optimization algorithm to harmonize the selected design variables. The results show that the weight and volume of the optimized ship steam turbine unit decrease by 4.2% and 6.3% respectively, which demonstrates the capability of the optimization method in optimizing the weight and volume of ship steam turbine unit.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130115195","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A robust design based on DFSS is presented for double wishbone suspension system kinematic and compliance (K&C) performance. Variations in suspension K&C caused by the uncertainties of hard points and bushing stiffness coefficients are minimized. The robust design involves two steps. In the first step, suspension kinematic characteristic are optimized. The objective functions are the toe angle and camber angle, and random design variables are the hardpoints of joints. The bushing stiffness coefficients are assumed as constant design parameters. In the second step, suspension compliance characteristics are optimized, where the bushing stiffness coefficients are random design variables. The optimized hardpoints in the first step are treated as constant design parameters. The optimization result shows that the robustness of suspension K&C performance is improved.
{"title":"Double Wishbone Suspension Design Based on Design for Six Sigma (DFSS)","authors":"Zhu Kaimin, Gu Jinxiang","doi":"10.1145/3421766.3421866","DOIUrl":"https://doi.org/10.1145/3421766.3421866","url":null,"abstract":"A robust design based on DFSS is presented for double wishbone suspension system kinematic and compliance (K&C) performance. Variations in suspension K&C caused by the uncertainties of hard points and bushing stiffness coefficients are minimized. The robust design involves two steps. In the first step, suspension kinematic characteristic are optimized. The objective functions are the toe angle and camber angle, and random design variables are the hardpoints of joints. The bushing stiffness coefficients are assumed as constant design parameters. In the second step, suspension compliance characteristics are optimized, where the bushing stiffness coefficients are random design variables. The optimized hardpoints in the first step are treated as constant design parameters. The optimization result shows that the robustness of suspension K&C performance is improved.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134345357","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The data mining algorithm based on rough set plays a very important role in dealing with various application-oriented problems. The suitable algorithm can quickly and accurately mine the core of time attribute and simplify the problem. Based on the characteristics of the teaching and appraise system of English linguistics, this paper optimizes the teaching design of English Linguistics in terms of teaching. Under the framework of systemic functional linguistics, this paper makes a follow-up analysis of the appraise resources in English texts, and verifies the feasibility and effectiveness of this method through an example. Some key problems of English linguistics teaching and price system are solved by data mining algorithm.
{"title":"Application of Data Mining in English Linguistics Teaching and Appraisal System","authors":"Wei Zhang, Xue Wang","doi":"10.1145/3421766.3421824","DOIUrl":"https://doi.org/10.1145/3421766.3421824","url":null,"abstract":"The data mining algorithm based on rough set plays a very important role in dealing with various application-oriented problems. The suitable algorithm can quickly and accurately mine the core of time attribute and simplify the problem. Based on the characteristics of the teaching and appraise system of English linguistics, this paper optimizes the teaching design of English Linguistics in terms of teaching. Under the framework of systemic functional linguistics, this paper makes a follow-up analysis of the appraise resources in English texts, and verifies the feasibility and effectiveness of this method through an example. Some key problems of English linguistics teaching and price system are solved by data mining algorithm.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134433965","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
CCTV News Broadcast is one of the most popular news programs in China, and it is also the most important propaganda platform in China. CCTV News Broadcast is established to "Improve the quality of publicity", so it is "A product of visual culture of national ideology" and "Taking politics as the standard" is primary appeal. [1] At present, there is little research on the text of CCTV News Broadcast. This paper focuses on the CCTV News Broadcast, using the visualization model based statistics and semantic based keyword extraction model (SKE) to extract the text features of CCTV News Broadcast. It can help the public quickly capture the key information of CCTV News Broadcast. Moreover, this paper also forms a set of Chinese corpus with keywords tagging in the field of CCTV News Broadcast. It provides important data support for machine learning method and subsequent research. In addition, aiming at some important problems found in this paper, this paper proposes further research direction for text data processing in CCTV News Broadcast field.
{"title":"CCTV News Broadcast Information Mining: Keyword Extraction Based on Semantic Model and Statistics Visualization","authors":"Yujie Xie, Fenghai Liu","doi":"10.1145/3421766.3421827","DOIUrl":"https://doi.org/10.1145/3421766.3421827","url":null,"abstract":"CCTV News Broadcast is one of the most popular news programs in China, and it is also the most important propaganda platform in China. CCTV News Broadcast is established to \"Improve the quality of publicity\", so it is \"A product of visual culture of national ideology\" and \"Taking politics as the standard\" is primary appeal. [1] At present, there is little research on the text of CCTV News Broadcast. This paper focuses on the CCTV News Broadcast, using the visualization model based statistics and semantic based keyword extraction model (SKE) to extract the text features of CCTV News Broadcast. It can help the public quickly capture the key information of CCTV News Broadcast. Moreover, this paper also forms a set of Chinese corpus with keywords tagging in the field of CCTV News Broadcast. It provides important data support for machine learning method and subsequent research. In addition, aiming at some important problems found in this paper, this paper proposes further research direction for text data processing in CCTV News Broadcast field.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114269767","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In the facial action unit (Facial Action Unit, AU) recognition process, due to the low occurrence probability of some AUs, the sample imbalance is serious, which severely limits the model recognition performance. Generative adversarial network GAN is an unsupervised learning method. Compared with the autoencoder and autoregressive model in the unsupervised learning method, its advantages are sufficient data fitting, higher efficiency and better generated samples. The original GAN model uses the minimum and maximum (minmax) to continuously optimize the training of the model; the conditional generation adversarial network CGAN adds conditional constraints to the model input to make the generated results controllable and prevent collapse in the model training process. GAN has been widely used in research fields such as image processing, natural language processing NLP, and real-time color correction of underwater images. This paper designs a model based on a conditional generation adversarial network to supplement the minority samples of a specific AU and improve the sample distribution space of the action unit.
在人脸动作单元(facial action unit, AU)识别过程中，由于某些人脸的出现概率较低，导致样本失衡严重，严重限制了模型的识别性能。生成对抗网络GAN是一种无监督学习方法。与无监督学习方法中的自编码器和自回归模型相比，其优点是数据拟合充分，效率更高，生成的样本质量更好。原始GAN模型采用最小最大值(minmax)对模型进行持续优化训练;条件生成对抗网络CGAN在模型输入中加入条件约束，使生成的结果可控，防止模型训练过程中的崩溃。GAN已广泛应用于图像处理、自然语言处理NLP、水下图像实时色彩校正等研究领域。本文设计了一种基于条件生成对抗网络的模型，以补充特定AU的少数样本，改善行动单元的样本分布空间。
{"title":"AU Data Augmentation Method Based on Generative Adversarial Networks","authors":"Qingdan Huang, Liqiang Pei, Yong Wang, Lian Zeng","doi":"10.1145/3421766.3421825","DOIUrl":"https://doi.org/10.1145/3421766.3421825","url":null,"abstract":"In the facial action unit (Facial Action Unit, AU) recognition process, due to the low occurrence probability of some AUs, the sample imbalance is serious, which severely limits the model recognition performance. Generative adversarial network GAN is an unsupervised learning method. Compared with the autoencoder and autoregressive model in the unsupervised learning method, its advantages are sufficient data fitting, higher efficiency and better generated samples. The original GAN model uses the minimum and maximum (minmax) to continuously optimize the training of the model; the conditional generation adversarial network CGAN adds conditional constraints to the model input to make the generated results controllable and prevent collapse in the model training process. GAN has been widely used in research fields such as image processing, natural language processing NLP, and real-time color correction of underwater images. This paper designs a model based on a conditional generation adversarial network to supplement the minority samples of a specific AU and improve the sample distribution space of the action unit.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"41 6 Suppl 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123682157","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Existing dehazing methods are usually to appear visual problems. In the paper, we put forward a truncated total variation method (TTV) to eliminate haze. A histogram analysis is firstly developed to obtain global atmospheric light. Then, using an adaptive boundary constraint TTV to optimize the transmission properly. Finally, a new DCP is presented to remove haze. Shown in experimental results, our method can outperform existent methods on the visual effect.
{"title":"Single Fog Image Dehazing via Truncated Total Variation Method","authors":"Yin Gao, Yijing Su, Jun Li","doi":"10.1145/3421766.3421772","DOIUrl":"https://doi.org/10.1145/3421766.3421772","url":null,"abstract":"Existing dehazing methods are usually to appear visual problems. In the paper, we put forward a truncated total variation method (TTV) to eliminate haze. A histogram analysis is firstly developed to obtain global atmospheric light. Then, using an adaptive boundary constraint TTV to optimize the transmission properly. Finally, a new DCP is presented to remove haze. Shown in experimental results, our method can outperform existent methods on the visual effect.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123754494","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents a novel design for a robotic end-effector. In particular, the design features a 4-DOF parallel gripper potential of applying for industrial automation. This gripper mainly consists of two parallelogram mechanisms and two grasp sliders, which can perform in-hand twisting action (in-hand manipulation) and in-plane horizontal and vertical transmission. Moreover, the proposed gripper adopts parallel grasping mode and can provide stable ability to maintain force closure on an object with large ranges of size. The kinematic analysis which includes position solution and velocity solution is derived. The velocity manipulability that possesses the physical significance of energy transfer efficiency is defined. A numerical example is presented to evaluate the performance of the gripper. The results indicate that the gripper can achieve horizontal transmission to supplement the workspace of the robot arm, and possess relatively better performance on in-hand manipulation and in-plane vertical transmission.
{"title":"Kinematic Analysis and Manipulability Evaluation of a 4-DOF Parallel Gripper for Dexterous Manipulation","authors":"Xiaodong Zhou, Jianfeng Li, Shiping Zuo","doi":"10.1145/3421766.3421861","DOIUrl":"https://doi.org/10.1145/3421766.3421861","url":null,"abstract":"This paper presents a novel design for a robotic end-effector. In particular, the design features a 4-DOF parallel gripper potential of applying for industrial automation. This gripper mainly consists of two parallelogram mechanisms and two grasp sliders, which can perform in-hand twisting action (in-hand manipulation) and in-plane horizontal and vertical transmission. Moreover, the proposed gripper adopts parallel grasping mode and can provide stable ability to maintain force closure on an object with large ranges of size. The kinematic analysis which includes position solution and velocity solution is derived. The velocity manipulability that possesses the physical significance of energy transfer efficiency is defined. A numerical example is presented to evaluate the performance of the gripper. The results indicate that the gripper can achieve horizontal transmission to supplement the workspace of the robot arm, and possess relatively better performance on in-hand manipulation and in-plane vertical transmission.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127883890","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A model renderer with style image generation is designed combine with illumination technology and style transfer technology for the illumination problems, that may be encountered in the process of computer image graphics design and production.In order to achieve a better rendering effect of the model, the rendering results are combined with style transfer innovatively.Different styles of model images can be applied in scenes such as games and movies to facilitate future development.
{"title":"Model Renderer Design with Style Image","authors":"Xiaozi Guo, Juan Zhang, Mingquan Zhou","doi":"10.1145/3421766.3421797","DOIUrl":"https://doi.org/10.1145/3421766.3421797","url":null,"abstract":"A model renderer with style image generation is designed combine with illumination technology and style transfer technology for the illumination problems, that may be encountered in the process of computer image graphics design and production.In order to achieve a better rendering effect of the model, the rendering results are combined with style transfer innovatively.Different styles of model images can be applied in scenes such as games and movies to facilitate future development.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"198 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115252528","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：