2019 IEEE International Conference on Cognitive Computing (ICCC)最新文献

Feature Importance Identification through Bottleneck Reconstruction 瓶颈重构特征重要性识别

2019 IEEE International Conference on Cognitive Computing (ICCC) Pub Date : 2019-07-08 DOI: 10.1109/ICCC.2019.00022

Linsong Chu, R. Raghavendra, M. Srivatsa, A. Preece, Daniel Harborne

引用次数: 0

The Visual Representation of Abstract Verbs: Merging Verb Classification with Iconicity in Sign Language 抽象动词的视觉表征:手语动词分类与象似性的融合

2019 IEEE International Conference on Cognitive Computing (ICCC) Pub Date : 2019-07-08 DOI: 10.1109/ICCC.2019.00025

Simone Scicluna, C. Strapparava

{"title":"The Visual Representation of Abstract Verbs: Merging Verb Classification with Iconicity in Sign Language","authors":"Simone Scicluna, C. Strapparava","doi":"10.1109/ICCC.2019.00025","DOIUrl":"https://doi.org/10.1109/ICCC.2019.00025","url":null,"abstract":"Theories like the picture superiority effect state that the visual modality has substantial advantage over the other human senses. This makes visual information vital in the acquisition of knowledge, such as in the learning of a language. Words can be graphically represented to illustrate the meaning of a message and facilitate its understanding. This method, however, becomes a limitation in the case of abstract words, like accept, belong, integrate and agree, which have no visual referent. The current research turns to sign languages to explore the common semantic elements that link words to each other. Such visual languages have been found to reveal enlightening patterns across signs of similar meanings, pointing towards the possibility of creating clusters of iconic meanings along with their respective graphic representation. By using sign language insight and VerbNet's organisation of verb predicates, this study presents a novel organisation of 506 English abstract verbs classified by visual shape. Graphic animation was used to visually represent the 20 classes of abstract verbs developed. To build confidence on the resulting product, which can be accessed on www.vroav.online, an online survey was created to achieve judgements on the visuals' representativeness. Considerable agreement between participants was found, suggesting a positive way forward for this work, which may be developed as a language learning aid in educational contexts or as a multimodal language comprehension tool for digital text.","PeriodicalId":262923,"journal":{"name":"2019 IEEE International Conference on Cognitive Computing (ICCC)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122024308","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}

引用次数: 1

Towards Classifying Motor Imagery Using a Consumer-Grade Brain-Computer Interface 使用消费级脑机接口对运动图像进行分类

2019 IEEE International Conference on Cognitive Computing (ICCC) Pub Date : 2019-07-08 DOI: 10.1109/ICCC.2019.00023

Ganyu Wang, Miguel Vargas Martin, P. Hung, Shane MacDonald

引用次数: 1

Tackling the Challenge of Computational Identification of Characters in Fictional Narratives 解决虚构叙事中人物计算识别的挑战

2019 IEEE International Conference on Cognitive Computing (ICCC) Pub Date : 2019-07-08 DOI: 10.1109/ICCC.2019.00031

Cristina Barros, Marta Vicente-Moreno, Elena Lloret

引用次数: 3

Identifying and Clustering Users for Unsupervised Intrusion Detection in Corporate Audit Sessions 企业审计会话中无监督入侵检测的用户识别与聚类

2019 IEEE International Conference on Cognitive Computing (ICCC) Pub Date : 2019-07-08 DOI: 10.1109/ICCC.2019.00016

Mathieu Garchery, M. Granitzer

引用次数: 2

Securing Malware Cognitive Systems against Adversarial Attacks 保护恶意软件认知系统免受对抗性攻击

2019 IEEE International Conference on Cognitive Computing (ICCC) Pub Date : 2019-07-08 DOI: 10.1109/ICCC.2019.00014

Yuede Ji, Benjamin Bowman, H. H. Huang

{"title":"Securing Malware Cognitive Systems against Adversarial Attacks","authors":"Yuede Ji, Benjamin Bowman, H. H. Huang","doi":"10.1109/ICCC.2019.00014","DOIUrl":"https://doi.org/10.1109/ICCC.2019.00014","url":null,"abstract":"The cognitive systems along with the machine learning techniques have provided significant improvements for many applications. However, recent adversarial attacks, such as data poisoning, evasion attacks, and exploratory attacks, have shown to be able to either cause the machine learning methods to misbehave, or leak sensitive model parameters. In this work, we have devised a prototype of a malware cognitive system, called DeepArmour, which performs robust malware classification against adversarial attacks. At the heart of our method is a voting system with three different machine learning malware classifiers: random forest, multi-layer perceptron, and structure2vec. In addition, DeepArmour applies several adversarial countermeasures, such as feature reconstruction and adversarial retraining to strengthen the robustness. We tested DeepArmour on a malware execution trace dataset, which has 12, 536 malware in five categories. We are able to achieve 0.989 accuracy with 10-fold cross validation. Further, to demonstrate the ability of combating adversarial attacks, we have performed a white-box evasion attack on the dataset and showed how our system is resilient to such attacks. Particularly, DeepArmour is able to achieve 0.675 accuracy for the generated adversarial attacks which are unknown to the model. After retraining with only 10% adversarial samples, DeepArmour is able to achieve 0.839 accuracy","PeriodicalId":262923,"journal":{"name":"2019 IEEE International Conference on Cognitive Computing (ICCC)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126293650","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}

引用次数: 16

The Impact of Peer Pressure: Extending Axelrod's Model on Cultural Polarisation 同伴压力的影响:扩展阿克塞尔罗德的文化两极分化模型

2019 IEEE International Conference on Cognitive Computing (ICCC) Pub Date : 2019-07-08 DOI: 10.1109/ICCC.2019.00030

Rhodri L. Morris, Liam D. Turner, R. Whitaker, Cheryl Giammanco

{"title":"The Impact of Peer Pressure: Extending Axelrod's Model on Cultural Polarisation","authors":"Rhodri L. Morris, Liam D. Turner, R. Whitaker, Cheryl Giammanco","doi":"10.1109/ICCC.2019.00030","DOIUrl":"https://doi.org/10.1109/ICCC.2019.00030","url":null,"abstract":"Culture represents the broad range of things over which people influence each other, and frequently contributes to the behaviour, interaction and outlook of groups. Although it has been studied in the context of humans, it is also relevant to future intelligent cognitive systems, that could have the capability to update their disposition and strategy based on the influence of others. In this work we transfer concepts from social sciences to the computing sciences and examine the effect of peer influence on culture. We consider the notion of \"peer pressure\", being the combined effect from all an individual's neighbours exerting influence at the same time, and also through influence flowing from indirect sources. This approach is derived using Social Impact Theory. We benchmark this against the cultural polarisation model from Axelrod, which involves influence being restricted to dyadic interactions between agents. We find that peer pressure provides complex contagion with a significant impact on cultural evolution. Greater cultural diversity is maintained, with indirect paths mitigating this by effectively forming disruptive weak links. This reaffirms that maintaining diversity in social ties, as well as a wide breadth, supports the mitigation of cultural isolation and polarisation. The model provides a platform to explore culture in a wide range of further scenarios, including electronic, coalition and organisational contexts.","PeriodicalId":262923,"journal":{"name":"2019 IEEE International Conference on Cognitive Computing (ICCC)","volume":"118 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132800582","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}

引用次数: 2

Scaling Deep Spiking Neural Networks with Binary Stochastic Activations 基于二元随机激活的深度尖峰神经网络

2019 IEEE International Conference on Cognitive Computing (ICCC) Pub Date : 2019-07-08 DOI: 10.1109/ICCC.2019.00020

Deboleena Roy, I. Chakraborty, K. Roy

{"title":"Scaling Deep Spiking Neural Networks with Binary Stochastic Activations","authors":"Deboleena Roy, I. Chakraborty, K. Roy","doi":"10.1109/ICCC.2019.00020","DOIUrl":"https://doi.org/10.1109/ICCC.2019.00020","url":null,"abstract":"The modern era has witnessed a proliferation of portable devices that use Artificial Intelligence (AI) to enhance user experiences. Majority of these AI tasks are performed by large neural networks, which require a good amount of memory and compute power. This has resulted in a growing interest in Spiking Neural Networks (SNNs) which communicate through binary activations or 'spikes', as they offer a bio-plausible and energy efficient alternative to traditional deep neural networks (DNNs). In this work, we present deep spiking neural networks with binary stochastic activations, that are tailored for implementation on emerging hardware platforms. We evaluate two deep neural network models, VGG-9 and VGG-16 on CIFAR-10 and CIFAR-100 datasets, respectively, with binary stochastic activations. We achieve state of the accuracy and achieve 1.4x improvement in energy consumption because of spike-based communication versus a network with ReLU neurons. We further investigate extremely quantized version of these networks having binary weights and show an energy benefit of 28x over full-precision neural networks. Thus we present scalable deep spiking neural networks that achieve performance comparable to DNNs while achieving substantial energy benefit.","PeriodicalId":262923,"journal":{"name":"2019 IEEE International Conference on Cognitive Computing (ICCC)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122062671","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}

引用次数: 20

Difficulties and Improvements to Graph-Based Lexical Sentiment Analysis Using LISA 基于LISA的基于图的词汇情感分析的难点与改进

2019 IEEE International Conference on Cognitive Computing (ICCC) Pub Date : 2019-07-08 DOI: 10.1109/ICCC.2019.00008

Mireille Fares, Angela Moufarrej, Eliane Jreij, Joe Tekli, W. Grosky

{"title":"Difficulties and Improvements to Graph-Based Lexical Sentiment Analysis Using LISA","authors":"Mireille Fares, Angela Moufarrej, Eliane Jreij, Joe Tekli, W. Grosky","doi":"10.1109/ICCC.2019.00008","DOIUrl":"https://doi.org/10.1109/ICCC.2019.00008","url":null,"abstract":"Lexical sentiment analysis (LSA) underlines a family of methods combining natural language processing, machine learning, or graph navigation techniques to identify the underlying sentiments or emotions carried in textual data. In this paper, we introduce LISA, an unsupervised word-level knowledge graph-based LexIcal Sentiment Analysis framework. It uses different variants of shortest path graph navigation techniques to compute and propagate affective scores in a lexical-affective graph (LAG), created by connecting a typical lexical knowledgebase (KB) like WordNet, with a reliable affect KB like WordNet-Affect Hierarchy. LISA was designed in two consecutive iterations, producing two main modules: i) LISA 1.0 for affect navigation, and ii) LISA 2.0 for affect propagation and lookup. LISA 1.0 suffered from the semantic connectivity problem shared by some existing lexicon-based methods, and required polynomial execution time. This led to the development of LISA 2.0, which i) processes affective relationships separately from lexical/semantic connections (solving the semantic connectivity problem of LISA 1.0), and ii) produces a sentiment lexicon which can be searched in logarithmic time (handling LISA 1.0's efficiency problem). Experimental results on the ANEW dataset show that LISA 2.0, while completely unsupervised, is on a par with existing supervised solutions, highlighting its quality and potential.","PeriodicalId":262923,"journal":{"name":"2019 IEEE International Conference on Cognitive Computing (ICCC)","volume":"437 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116554096","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}

引用次数: 7

Extraction of Taxonomic Relation of Complex Terms by Recurrent Neural Network 复项分类关系的递归神经网络提取

2019 IEEE International Conference on Cognitive Computing (ICCC) Pub Date : 2019-07-08 DOI: 10.1109/ICCC.2019.00024

Atsushi Oba, Incheon Paik

引用次数: 1