2019 IEEE First International Conference on Cognitive Machine Intelligence (CogMI)最新文献

{"title":"Deployment of Deep Learning Models to Mobile Devices for Spam Classification","authors":"Ameema Zainab, Dabeeruddin Syed, Dena Al-Thani","doi":"10.1109/CogMI48466.2019.00024","DOIUrl":"https://doi.org/10.1109/CogMI48466.2019.00024","url":null,"abstract":"The advent of deep learning brings the possibility of better and faster applications in real world. In this work, deep learning models are used for application of spam classification in mobile devices. A Binary Classification model is trained with deep learning and is transformed to a graph using tensorflow and then, is converted to a protobuf file to be deployed on mobile devices. Instead of looking into the spam messages in an algorithmic way i.e. just with keywords, binary model deals with experience of learning and predicts if a text message is spam. The training was performed multiple times on resource-deficient devices and hyper-parameter optimization was performed to enhance the training accuracy to 99.87 %. The test accuracy of mobile application is 98.7 % and testing happens in real-time without any internet access. Our simulation shows that a model with an embedding layer (size 128), an LSTM layer (size 64, dropout 0.2) and a dense layer (sigmoid) yields the highest performance. Also, the comparative evaluation with state-of-the-art methods displayed that our model achieves higher accuracy.","PeriodicalId":116160,"journal":{"name":"2019 IEEE First International Conference on Cognitive Machine Intelligence (CogMI)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125587975","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}

{"title":"Cognitive Identity Management: Risks, Trust and Decisions using Heterogeneous Sources","authors":"S. Yanushkevich, W. Howells, Keeley A. Crockett, J. O'Shea, H. C. R. Oliveira, R. Guest, V. Shmerko","doi":"10.1109/CogMI48466.2019.00014","DOIUrl":"https://doi.org/10.1109/CogMI48466.2019.00014","url":null,"abstract":"This work advocates for cognitive biometric-enabled systems that integrate identity management, risk assessment and trust assessment. The cognitive identity management process is viewed as a multi-state dynamical system, and probabilistic reasoning is used for modeling of this process. This paper describes an approach to design a platform for risk and trust modeling and evaluation in the cognitive identity management built upon processing heterogeneous data including biometrics, other sensory data and digital ID. The core of an approach is the perception-action cycle of each system state. Inference engine is a causal network that uses various uncertainty metrics and reasoning mechanisms including Dempster-Shafer and Dezert-Smarandache beliefs.","PeriodicalId":116160,"journal":{"name":"2019 IEEE First International Conference on Cognitive Machine Intelligence (CogMI)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124561419","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}

{"title":"Approximate Deep Network Embedding for Mining Large-Scale Graphs","authors":"Yang Zhou, Ling Liu","doi":"10.1109/CogMI48466.2019.00016","DOIUrl":"https://doi.org/10.1109/CogMI48466.2019.00016","url":null,"abstract":"In the Big Data era, we are witnessing the flood of big graph data in terms of volume, variety, and velocity. The use of online social media and online shopping sites has provided access to a huge volume of interactions among information entities and objects. Scaling compute-intensive graph analysis applications on huge graphs with millions or billions of vertices and edges is widely recognized as a challenging big data research problem. A lot of research efforts in network embedding aim to learn low-dimensional representation of big graphs, enable easy integration with existing graph mining algorithms, and thus allow to achieve acceptable quality of big graph analysis on network embedding results with superior efficiency and scalability. However, how to enhance network embedding itself in terms of both efficiency and scalability is still an open problem. We are still short of efficient and scalable network embedding techniques to scale themselves on big graphs with millions or billions of vertices and edges, with the awareness of the intrinsic global and local characteristics of graph data. Most network embedding techniques exploit shallow-structured architectures, and thus lead to sub-optimal network representations. We also see lots of potential to utilize approximation theories and deep learning techniques to elevate both efficiency and scalability. In order to promote big network embedding from theoretical points of view, by representing graph data in deep learning architectures, we develop a suite of competitive learning-based approximate deep network embedding techniques that are able to leverage both efficiency and scalability of network embedding while preserving the computational utility with three major components. First, we propose a dynamic competitive learning-based algorithm to combine global network embedding and local network embedding into a unified model to utilize the advantages of both techniques. Second, we develop a network embedding-based algorithm with the optimization of competitive learning to tightly integrate vertex clustering and edge clustering by mutually enhancing each other. Third but last, we explore the opportunities of competitive learning and ranking for the optimal top-K neuron selection in the learning process of deep network embedding, in order to achieve a good balance between effectiveness and efficiency. The approximate deep network embedding approaches allow the deep learning model themselves to deactivate those insignificant neurons in the hidden layers through competitive learning, and thus reduce the computational cost of the feedforward pass and the back propagation.","PeriodicalId":116160,"journal":{"name":"2019 IEEE First International Conference on Cognitive Machine Intelligence (CogMI)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123502270","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}

{"title":"Multimodal Models for Contextual Affect Assessment in Real-Time","authors":"J. Vice, M. Khan, S. Yanushkevich","doi":"10.1109/CogMI48466.2019.00020","DOIUrl":"https://doi.org/10.1109/CogMI48466.2019.00020","url":null,"abstract":"Most affect classification schemes rely on near accurate single-cue models resulting in less than required accuracy under certain peculiar conditions. We investigate how the holism of a multimodal solution could be exploited for affect classification. This paper presents the design and implementation of a prototype, stand-alone, real-time multimodal affective state classification system. The presented system utilizes speech and facial muscle movements to create a holistic classifier. The system combines a facial expression classifier and a speech classifier that analyses speech through paralanguage and propositional content. The proposed classification scheme includes a Support Vector Machine (SVM) - paralanguage; a K-Nearest Neighbor (KNN) - propositional content and an InceptionV3 neural network - facial expressions of affective states. The SVM and Inception models boasted respective validation accuracies of 99.2% and 92.78%.","PeriodicalId":116160,"journal":{"name":"2019 IEEE First International Conference on Cognitive Machine Intelligence (CogMI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131188257","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}

{"title":"Sensory Audio Focusing Detection Using Brain-Computer Interface Archetype","authors":"Ryan Villanueva, Brandon Hoang, Urmil Shah, Yazmin Martinez, K. George","doi":"10.1109/CogMI48466.2019.00022","DOIUrl":"https://doi.org/10.1109/CogMI48466.2019.00022","url":null,"abstract":"Everyday people are placed in environments where countless conversations simultaneously take place within earshot. Speech intelligibility in the presence of multiple speakers, commonly known as the 'Cocktail Party Phenomenon', is significantly reduced for most hearing-impaired listeners who use hearing assistive devices [1]. Prior research addressing this issue include noise filtering based on trajectories of multiple moving speakers and locations of talking targets based on face detection [2][3]. This study focuses on the practicality of audio filtering through measuring electroencephalogram (EEG) signals using a Brain-Computer Interfaces (BCI) system. The study explores the use of machine learning algorithms to classify which speaker the listener is focusing on. In this study, training data is obtained of a listener focusing on one auditory stimulus (audiobook) while other auditory stimuli are presented at the same time. A g.Nautilus BCI headset was used to obtain EEG data. After collecting trial data for each audio source, a machine learning algorithm trains a classifier to distinguish one audiobook between another. Data was collected from five subjects in each trial. Results yielded an accuracy of above 90% from all three experiments.","PeriodicalId":116160,"journal":{"name":"2019 IEEE First International Conference on Cognitive Machine Intelligence (CogMI)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125079501","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}

{"title":"Building Explainable Predictive Analytics for Location-Dependent Time-Series Data","authors":"Yao-Yi Chiang, Yijun Lin, M. Franklin, S. Eckel, J. Ambite, Wei-Shinn Ku","doi":"10.1109/CogMI48466.2019.00037","DOIUrl":"https://doi.org/10.1109/CogMI48466.2019.00037","url":null,"abstract":"There are increasing numbers of online sources of real-time and historical location-dependent time-series data describing various types of environmental phenomena, e.g., traffic conditions and air quality levels. When coupled with the information that characterizes the natural and built environments, these location-dependent time-series data can help better understand interactions between and within human social systems and the ecosystem. Nevertheless, these data are still limited by their spatial and temporal resolution for downstream use (e.g., generating residential-level environmental exposures for human health studies). In this paper, we present a vision of a general machine learning framework for explainable predictive analytics for location-dependent time-series data. The framework will effectively deal with data-and model-related challenges for general scientific predictive analytics on spatiotemporal environmental phenomena. The challenges include how to identify the main features driving the phenomena, how to handle complex spatiotemporal variations in the phenomena, and how to utilize sparse ground truth measurements for training and validation. The resulting framework will enable fine spatial and temporal scale environmental exposure assessment and allow researchers to carry out unprecedented inquiries, such as understanding relationships between health outcomes and long-term air pollution exposures.","PeriodicalId":116160,"journal":{"name":"2019 IEEE First International Conference on Cognitive Machine Intelligence (CogMI)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128916619","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}

{"title":"GRAHIES: Multi-Scale Graph Representation Learning with Latent Hierarchical Structure","authors":"Lei Yu, Qi Zhang, Donna E. Dillenberger, Ling Liu, C. Pu, K. Chow, M. E. Gursoy, Stacey Truex, Hong Min, A. Iyengar, Gong Su","doi":"10.1109/CogMI48466.2019.00011","DOIUrl":"https://doi.org/10.1109/CogMI48466.2019.00011","url":null,"abstract":"A wide variety of deep neural network models for graph-structured data have been proposed to solve tasks like node/graph classification and link prediction. By effectively learning low-dimensional embeddings of graph nodes, they have shown state-of-the-art performance. However, most existing models learn node embeddings by exploring flat information propagation across the edges within the local neighborhood of each node. We argue that incorporating hierarchical node embeddings can capture the inherently hierarchical topological features of many realistic graphs such as social networks, biological network and World Wide Web. In this paper we propose GRAHIES, a general framework for graph neural networks to learn node representations that preserve hierarchical graph information at higher-orders. GRAHIES adaptively learns a multi-level hierarchical structure of the input graph, which consists of successively coarser (smaller) graphs that preserve the global structure of the original graphs at different levels. By combining the graph representations from different levels of the graph hierarchy, the final node representation captures the inherent global hierarchical structure of the original graph. Our experiments show that applying GRAHIES's hierarchical paradigm yields improved accuracy for existing graph neural networks on the node classification tasks.","PeriodicalId":116160,"journal":{"name":"2019 IEEE First International Conference on Cognitive Machine Intelligence (CogMI)","volume":"71 7","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131894176","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}

{"title":"Applying Software Design Metrics to Developer Story: A Supervised Machine Learning Analysis","authors":"Asaad Algarni, Kenneth I. Magel","doi":"10.1109/CogMI48466.2019.00030","DOIUrl":"https://doi.org/10.1109/CogMI48466.2019.00030","url":null,"abstract":"Object-oriented analysis is a significant step that plays a vital role in the success of software development. The planning and management stages, in particular, profoundly rely on the deliverance of an accurate estimate that takes the software's complexity and size into consideration. Today, several software industries are transforming their development methodologies to Agile due to its ability to deliver value in a short time and its cost efficiency. However, Agile methods prevent heavyweight modeling analysis and depend on user stories to drive the estimation process. Because user stories are descriptive language, they may not provide a clear picture for the implementation. Also, they may not help Agile developers give an accurate estimation due to their difficulty in measuring the complexity and size of a feature. Thus, this paper presents a new Agile artifact called developer story that allows the Agile developer to not only analyze and design software products but also predict the size of each feature, including its complexity. In this paper, a case study is presented that shows how the utilization of developer story is a practical approach in predicting the source code size of a feature and its complexity.","PeriodicalId":116160,"journal":{"name":"2019 IEEE First International Conference on Cognitive Machine Intelligence (CogMI)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122582170","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}

{"title":"On Urban Event Tracking from Online Media: A Social Cognition Perspective","authors":"T. Abdelzaher","doi":"10.1109/CogMI48466.2019.00031","DOIUrl":"https://doi.org/10.1109/CogMI48466.2019.00031","url":null,"abstract":"Modern online media, such as Twitter, Instagram, and YouTube, democratized information broadcast, allowing anyone to offer content for large-scale dissemination. The resulting global accessibility of real-time information marked an unprecedented change in human history. It ushered-in an age of information overload and introduced fundamentally new content dissemination dynamics within a very short period of time compared to that necessary for our cognitive faculties to co-evolve. In the meantime, the public nature of offered information allows automated tools to observe not only what is being transmitted but also how it propagates. Actual diffusion of information is driven by recepients, who must individually prioritize what to consume and propagate in the face of mounting overload. The resulting propagation patterns offer insights into the collective cognitive choices made by the underlying population. Automated algorithms can harvest these insights to offer added value to the population at hand. One example is content curation (or recommendation) services that help users sift through increasingly larger amounts of information clutter to find the most relevant and interesting items. We argue that contemporary information distillation services that manage overload can lead to significant negative side-effects that may range from unintentional suppression of pertinent information to the undermining of the very foundations of modern democracy. This paper explains the mechanism by which these side effects occur and explores possible research directions surrounding the mitigation of such side effects, set in the context of urban event tracking applications.","PeriodicalId":116160,"journal":{"name":"2019 IEEE First International Conference on Cognitive Machine Intelligence (CogMI)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123914616","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}

{"title":"[Title page i]","authors":"","doi":"10.1109/cogmi48466.2019.00001","DOIUrl":"https://doi.org/10.1109/cogmi48466.2019.00001","url":null,"abstract":"","PeriodicalId":116160,"journal":{"name":"2019 IEEE First International Conference on Cognitive Machine Intelligence (CogMI)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130839274","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}