Structure “Research” on Academic Website ● 8 research areas (same as key expertise on industrial site, see below) ● x research topics per research area o these correspond to a ZAP team, part of a ZAP team, or the union of (parts of) ZAP teams o (Part of) a ZAP (team) can be in more than one research topic Template per research topic ● Research area (make a choice) o Distributed intelligence in IoT o Machine Learning & Data Mining o Semantic Intelligence o Multimedia Coding and Delivery o Communication Sub-Systems o Wireless Networking o Fixed Networking o Cloud & Big Data Infrastructures o OTHER ● Title o shorter = better (cf. links and breadcrumbs) ● Text describing the research topic o 2 – 4 paragraphs is okay o example below is quite long and will be shortened… ● Staff o list of names ● Researchers o list of names ● Projects (optional) o list of projects, possibly with a link to external project page o please provide funding channel, acronym, link, and title ● Key publications o ca. 3-7 key publications: please provide a bullet list of references and the biblio ID per publication ● Pictures o 1-3 relevant and representative (appealing) pictures (possibly with a caption) o also send pictures as attachments to email (to avoid problems extracting pictures from Word). RESEARCH TRACK 1: ● Research area (make a choice) Semantic Intelligence ● Title o shorter = better (cf. links and breadcrumbs) Multimedia Content Analysis ● Text describing the research topic o 2 – 4 paragraphs is okay o example below is quite long and will be shortened… The past few years have witnessed a significant increase in both the consumption and availability of online multimedia content. This can be attributed to the introduction of easy-to-use devices and online services, the availability of cheap storage and bandwidth, and more and more people going online. To give a few statistics, as of January 2015, people watch six billion hours of video content each month on YouTube, and upload 300 hours of video content each minute. Similarly, 300 to 500 million microposts are created on Twitter on a daily basis. These statistics are indicative of the problem of multimedia content overload (infobesity): our ability to manage and consume multimedia content is not able to keep up with our ability to create multimedia content. Our main research objective is to narrow the gap between multimedia content creation on the one hand, and multimedia content management and consumption on the other hand. To that end, we focus on the development of novel techniques for machine-based understanding of both textual and visual content, paying particular attention to the use of deep learning. The term ‘deep learning’ was coined in 2006, and refers to machine learning algorithms that make use of multiple non-linear layers to construct feature hierarchies, typically through the use of artificial neural networks. Specifically, our research efforts focus on the following topics: ● Machine learning for modeling and understanding of natural language ● Natural language processing of noisy and short form text, such as status updates on social networks ● Machine learning for modeling and understanding of visual data, such as news broadcast video clips and short-form noisy video clips ● Staff o list of names Wesley De Neve, Azarakhsh Jalalvand, Joni Dambre. ● Researchers o list of names Fréderic Godin, Baptist Vandersmissen. ● Projects (optional) o list of projects, possibly with a link to external project page o please provide funding channel, acronym, link, and title − MiX-ICON STEAMER (https://www.iminds.be/en/projects/steamer): Smart Text Enrichment Algorithms for MEdia Retrieval applications − MiX-ICON Audience Measurement (https://www.iminds.be/en/projects/METEN-KRIJTLIJNEN): Research into new measurement protocols and multi-platform media consumption − ICON iRead+ (https://www.iminds.be/en/projects/iread): The Intelligent Reading Companion ● Key publications o ca. 3-7 key publications: please provide a bullet list of references and the biblio ID per publication IDs: − https://biblio.ugent.be/publication/8048654 − https://biblio.ugent.be/publication/5939109 − https://biblio.ugent.be/publication/7033443 − https://biblio.ugent.be/publication/5939119 − https://biblio.ugent.be/publication/5817852 − https://biblio.ugent.be/publication/5732010 − https://biblio.ugent.be/publication/5817855 − Baptist Vandersmissen, Lucas Sterckx, Thomas Demeester, Azarakhsh Jalalvand, Wesley De Neve and Rik Van de Walle (2016). An automated end-to-end pipeline for fine-grained video annotation using deep neural networks. International Conference on Multimedia Retrieval. p. 409-412 − Fréderic Godin, Wesley De Neve and Rik Van de Walle (2015). Part-ofspeech tagging of Twitter microposts only using distributed word representations and a neural network. Computational linguistics in the Netherlands (CLIN 2015). p. 45 − Fréderic Godin, Baptist Vandersmissen, Wesley De Neve and Rik Van de Walle (2015). Multimedia Lab @ ACL W-NUT NER shared task: named entity recognition for Twitter microposts using distributed word representations. ACL 2015 Workshop on Noisy User-generated Text, p. 146-153 − Hans Paulussen, Francisco Capdevila, Pedro Debevere, Maribel M. Perez, Martin Vanbrabant, Wesley De Neve and Stefan De Wannemacker (2014). Building an NLP pipeline within a digital publishing workflow. Computational Linguistics in the Netherlands Journal. p. 71-84 − Fréderic Godin, Jasper Zuallaert, Baptist Vandersmissen, Wesley De Neve and Rik Van de Walle (2014). Beating the bookmakers: leveraging statistics and Twitter microposts for predicting soccer results. Workshop on Large-Scale Sports Analytics − Baptist Vandersmissen, Fréderic Godin, Abhineshwar Tomar, Wesley De Neve and Rik Van de Walle (2014). The rise of mobile and social short-form video: an in-depth measurement study of Vine. CEUR workshop proceedings. 1198. p. 1-10 − Fréderic Godin, Baptist Vandersmissen, Azarakhsh Jalalvand, Wesley De Neve and Rik Van de Walle (2014). Alleviating manual feature engineering for part-of-speech tagging of Twitter microposts using distributed word representations. NIPS Workshop on Modern Machine Learning and Natural Language Processing ● Pictures o 1-3 relevant and representative (appealing) pictures (possibly with a caption) o also send pictures as attachments to email (to avoid problems extracting pictures from Word). Visualization of a word vector space of the 1000 most frequent words on Twitter. During training, the model learned which words are similar to each other. For example, the word “the” is similar to the slang word “da” on Twitter. Visualization of the automatic detection of products in a YouTube video clip and linkage to a retailer website. Machine learning models learn to recognize and link products in video clips to facilitate e-commerce. RESEARCH TRACK 2: ● Research area (make a choice) Machine Learning & Data Mining ● Title ○ shorter = better (cf. links and breadcrumbs) Biotech Data Processing and Analysis ● Text describing the research topic ○ 2 – 4 paragraphs is okay ○ example below is quite long and will be shortened… Recent technological advances have led to the generation of vast collections of biotech data. As an example, the European Bioinformatics Institute (EBI) currently stores 20 petabytes of data and back-ups about genes, proteins, and small molecules. This data avalanche has created a strong need for novel mathematical and computational techniques to address challenges related to disease and our environment. In our research, we explore how multimedia technology can be used to process and analyze biotech data. In that regard, we have two major research objectives. A first objective is to leverage state-of-the-art video compression technology for the compression of genomics data, so to mitigate issues in terms of storage, transportation, and analysis. A second objective is to leverage deep machine learning in the context of several biotech use cases, ranging from genome annotation (e.g., splice site detection in DNA sequences) to tumor detection in medical images. Of particular interest is the application of techniques for natural language understanding to the analysis of genomics data. ● Staff ○ list of names Wesley De Neve, Joni Dambre ● Researchers ○ list of names Tom Paridaens, Jasper Zuallaert, Mijung Kim, Lionel Pigou ● Projects (optional) ○ list of projects, possibly with a link to external project page ○ please provide funding channel, acronym, link, and title N/A. ● Key publications ○ ca. 3-7 key publications: please provide a bullet list of references and the biblio ID per publication IDs: − https://biblio.ugent.be/publication/7198983 − https://biblio.ugent.be/publication/5821901 − https://biblio.ugent.be/publication/4378035 − Tom Paridaens, Jens Panneel, Wesley De Neve, Peter Lambert and Rik Van de Walle (2016). Leveraging CABAC for no-reference compression of genomic data with random access support. Data Compression Conference (DCC). − Tom Paridaens, Yves Van Stappen, Wesley De Neve, Peter Lambert and Rik Van de Walle (2014). Towards block-based compression of genomic data with random access functionality. Workshop on Genomic Signal Processing and Statistics. p. 1528-1531 − Tom Paridaens, Wesley De Neve, Peter Lambert and Rik Van de Walle (2014). Genome sequences as media files. 8th International joint conference on biomedical engineering systems and technologies. · Pictures o 1-3 relevant and representative (appealing) pictures (possibly with a caption) o also send pictures as attachments to email (to avoid problems extracting pictures from Word). The proposed solution for DNA compression allows for a flexible trade-off between efficiency, effectiveness, and functionality, setting our solution apart from the state-of-the-art. A deep neural network has been trained to recognize whether a given image contains a malignant lesion. When this is the case, the deep neural network subsequently localizes and extracts the malignant lesion. Automatic genome annotation, such as predicting the location of splice sites, translation initiation sites, or secondary structures. By detecting indicative patterns in an exemplary set of DNA sequences, machine learning models can make predictions for new DNA sequences.