X-Linear Attention Networks for Image Captioning Pan, Yingwei; Yao, Ting; Li, Yehao ...
2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR),
06/2020
Conference Proceeding
Odprti dostop
Recent progress on fine-grained visual recognition and visual question answering has featured Bilinear Pooling, which effectively models the 2nd order interactions across multi-modal inputs. ...Nevertheless, there has not been evidence in support of building such interactions concurrently with attention mechanism for image captioning. In this paper, we introduce a unified attention block --- X-Linear attention block, that fully employs bilinear pooling to selectively capitalize on visual information or perform multi-modal reasoning. Technically, X-Linear attention block simultaneously exploits both the spatial and channel-wise bilinear attention distributions to capture the 2^{nd} order interactions between the input single-modal or multi-modal features. Higher and even infinity order feature interactions are readily modeled through stacking multiple X-Linear attention blocks and equipping the block with Exponential Linear Unit (ELU) in a parameter-free fashion, respectively. Furthermore, we present X-Linear Attention Networks (dubbed as X-LAN) that novelly integrates X-Linear attention block(s) into image encoder and sentence decoder of image captioning model to leverage higher order intra- and inter-modal interactions. The experiments on COCO benchmark demonstrate that our X-LAN obtains to-date the best published CIDEr performance of 132.0% on COCO Karpathy test split. When further endowing Transformer with X-Linear attention blocks, CIDEr is boosted up to 132.8%. Source code is available at https://github.com/Panda-Peter/image-captioning.
CRISP-DM(CRoss-Industry Standard Process for Data Mining) has its origins in the second half of the nineties and is thus about two decades old. According to many surveys and user polls it is still ...the de facto standard for developing data mining and knowledge discovery projects. However, undoubtedly the field has moved on considerably in twenty years, with data science now the leading term being favoured over data mining . In this paper we investigate whether, and in what contexts, CRISP-DM is still fit for purpose for data science projects. We argue that if the project is goal-directed and process-driven the process model view still largely holds. On the other hand, when data science projects become more exploratory the paths that the project can take become more varied, and a more flexible model is called for. We suggest what the outlines of such a trajectory-based model might look like and how it can be used to categorise data science projects (goal-directed, exploratory or data management). We examine seven real-life exemplars where exploratory activities play an important role and compare them against 51 use cases extracted from the NIST Big Data Public Working Group. We anticipate this categorisation can help project planning in terms of time and cost characteristics.
Data Mining; Text Mining; Health Informatics; Health Care Information Systems; Medical Terminologies; Natural Language Processing; Text Analysis; Support Vector Machines
This open access book provides an overview of the recent advances in representation learning theory, algorithms and applications for natural language processing (NLP). It is divided into three parts. ...Part I presents the representation learning techniques for multiple language entries, including words, phrases, sentences and documents. Part II then introduces the representation techniques for those objects that are closely related to NLP, including entity-based world knowledge, sememe-based linguistic knowledge, networks, and cross-modal entries. Lastly, Part III provides open resource tools for representation learning techniques, and discusses the remaining challenges and future research directions. The theories and algorithms of representation learning presented can also benefit other related domains such as machine learning, social network analysis, semantic Web, information retrieval, data mining and computational biology. This book is intended for advanced undergraduate and graduate students, post-doctoral fellows, researchers, lecturers, and industrial engineers, as well as anyone interested in representation learning and natural language processing.
Data Mining: Practical Machine Learning Tools and Techniques, Third Edition, offers a thorough grounding in machine learning concepts as well as practical advice on applying machine learning tools ...and techniques in real-world data mining situations. This highly anticipated third edition of the most acclaimed work on data mining and machine learning will teach you everything you need to know about preparing inputs, interpreting outputs, evaluating results, and the algorithmic methods at the heart of successful data mining. Thorough updates reflect the technical changes and modernizations that have taken place in the field since the last edition, including new material on Data Transformations, Ensemble Learning, Massive Data Sets, Multi-instance Learning, plus a new version of the popular Weka machine learning software developed by the authors. Witten, Frank, and Hall include both tried-and-true techniques of today as well as methods at the leading edge of contemporary research. The book is targeted at information systems practitioners, programmers, consultants, developers, information technology managers, specification writers, data analysts, data modelers, database R professionals, data warehouse engineers, data mining professionals. The book will also be useful for professors and students of upper-level undergraduate and graduate-level data mining and machine learning courses who want to incorporate data mining as part of their data management knowledge base and expertise. * Provides a thorough grounding in machine learning concepts as well as practical advice on applying the tools and techniques to your data mining projects * Offers concrete tips and techniques for performance improvement that work by transforming the input or output in machine learning methods * Includes downloadable Weka software toolkit, a collection of machine learning algorithms for data mining tasks—in an updated, interactive interface. Algorithms in toolkit cover: data pre-processing, classification, regression, clustering, association rules, visualization
This open access book includes methods for retrieval, semantic representation, and analysis of Volunteered Geographic Information (VGI), geovisualization and user interactions related to VGI, and ...discusses selected topics in active participation, social context, and privacy awareness. It presents the results of the DFG-funded priority program "VGI: Interpretation, Visualization, and Social Computing" (2016-2023). The book includes three parts representing the principal research pillars within the program. Part I "Representation and Analysis of VGI" discusses recent approaches to enhance the representation and analysis of VGI. It includes semantic representation of VGI data in knowledge graphs; machine-learning approaches to VGI mining, completion, and enrichment as well as to the improvement of data quality and fitness for purpose. Part II "Geovisualization and User Interactions related to VGI" book explores geovisualizations and user interactions supporting the analysis and presentation of VGI data. When designing these visualizations and user interactions, the specific properties of VGI data, the knowledge and abilities of different target users, and technical viability of solutions need to be considered. Part III "Active Participation, Social Context and Privacy Awareness" of the book addresses the human impact associated with VGI. It includes chapters on the use of wearable sensors worn by volunteers to record their exposure to environmental stressors on their daily journeys, on the collective behavior of people using location-based social media and movement data from football matches, and on the motivation of volunteers who provide important support in information gathering, filtering and analysis of social media in disaster situations. The book is of interest to researchers and advanced professionals in geoinformation, cartography, visual analytics, data science and machine learning.
This open access comprehensive methodological book offers a much-needed answer to the lack of resources and methodological guidance in learning analytics, which has been a problem ever since the ...field started. The book covers all important quantitative topics in education at large as well as the latest in learning analytics and education data mining. The book also goes deeper into advanced methods that are at the forefront of novel methodological innovations. Authors of the book include world-renowned learning analytics researchers, R package developers, and methodological experts from diverse fields offering an unprecedented interdisciplinary reference on novel topics that is hard to find elsewhere. The book starts with the basics of R as a programming language, the basics of data cleaning, data manipulation, statistics, and analytics. In doing so, the book is suitable for newcomers as they can find an easy entry to the field, as well as being comprehensive of all the major methodologies. For every method, the corresponding chapter starts with the basics, explains the main concepts, and reviews examples from the literature. Every chapter has a detailed explanation of the essential techniques and basic functions combined with code and a full tutorial of the analysis with open-access real-life data. A total of 22 chapters are included in the book covering a wide range of methods such as predictive learning analytics, network analysis, temporal networks, epistemic networks, sequence analysis, process mining, factor analysis, structural topic modeling, clustering, longitudinal analysis, and Markov models. What is really unique about the book is that researchers can perform the most advanced analysis with the included code using the step-by-step tutorial and the included data without the need for any extra resources. This is an open access book.
The contributions gathered in this open access book focus on modern methods for data science and classification and present a series of real-world applications. Numerous research topics are covered, ...ranging from statistical inference and modeling to clustering and dimension reduction, from functional data analysis to time series analysis, and network analysis. The applications reflect new analyses in a variety of fields, including medicine, marketing, genetics, engineering, and education. The book comprises selected and peer-reviewed papers presented at the 17th Conference of the International Federation of Classification Societies (IFCS 2022), held in Porto, Portugal, July 19–23, 2022. The IFCS federates the classification societies and the IFCS biennial conference brings together researchers and stakeholders in the areas of Data Science, Classification, and Machine Learning. It provides a forum for presenting high-quality theoretical and applied works, and promoting and fostering interdisciplinary research and international cooperation. The intended audience is researchers and practitioners who seek the latest developments and applications in the field of data science and classification.