/news-archive/tag/social-computing/ 糖心Vlog传媒 Little Rock Wed, 01 Sep 2021 13:56:06 +0000 en-US hourly 1 https://wordpress.org/?v=6.9.4 /news-archive/2021/09/01/researchers-earn-best-paper-award/ Wed, 01 Sep 2021 13:56:06 +0000 /news/?p=78299 ... 糖心Vlog传媒 Little Rock researchers earn Best Paper Award for research using moviebarcode as video analysis method]]> A group of researchers from the University of Arkansas at Little Rock Collaboratorium for Social Media and Online Behavioral Studies (COSMOS) have been recognized for their research using moviebarcode, a technique used to summarize videos by compressing an entire video into a single image, to systematically categorize and analyze videos on YouTube. COSMOS researchers earned the 鈥�鈥� for their submission, 鈥�,鈥� at the 听held in Porto, Portugal.听 The paper was written by , Rick Rejeleene, and, doctoral students in computer and information science at 糖心Vlog传媒 Little Rock, , a 糖心Vlog传媒 Little Rock graduate, Dr. , post-doctoral researcher at COSMOS, and Dr. , Jerry L. Maulden-Entergy endowed chair and distinguished professor of information science and director of COSMOS.听 鈥淢oviebarcodes are typically used to visualize a summary of videos,鈥� said Erol, the lead author. 鈥淗owever, we used the color theory computationally so that we can categorize videos using this technology. This technique compresses a video into an image, then it summarizes the video. The computation time is very small compared to other deep learning-based algorithms.鈥� Having a tool that allows a quicker method to analyze videos without watching them is important for researchers who study online videos. YouTube, for example, is a popular source of videos for researchers. Approximately one billion hours of video are added to YouTube daily. 鈥淐osmographers have a knack of thinking out of the box and producing groundbreaking innovations,鈥� Dr. Agarwal said. 鈥淢oviebarcode-based video data analysis is one such idea with numerous applications that COSMOS is developing including faster and efficient search and retrieval, computationally extracting noteworthy narrative elements in videos, identifying coordinated influence campaigns, among others.鈥� The research idea stemmed from a project Young worked on during the spring 2019 Social Computing course offered by Dr. Agarwal. 鈥淲hat I find interesting about moviebarcodes is that hours of video content can be quickly digested by a user in a single glance,鈥� Young stated. 鈥淎fter researching moviebarcodes further, I began considering possible applications to my work with COSMOS proposing the development of a moviebarcodes tool that can effectively visualize video data and provide impactful insights.鈥� Each barcode consists of generated colors for every frame of the movie making it unique. All frames combined into one code showed color transition within videos and allowed for a comparison with other videos as well as a highly accurate grouping of videos without having to watch them.听 The team was also able to reduce time and computing energy required to categorize videos by creating video collections and then categorizing videos using a clustering method. 鈥淭his is a big advantage on video processing. By just looking at moviebarcodes, one can easily find what kind of video it is,鈥� Erol said. 鈥�Nowadays, deep learning-based algorithms and tools are used for similar purposes. However, video processing requires a huge amount of GPU computation power and resources. Not all researchers may have access to these resources. With our research, we try to minimize computing requirements for large scale video processing and make this tool available for all users.鈥� The COSMOS team is already exploring how moviebarcode can be used for different research applications. 鈥淲e are using the moviebarcode technique for different research areas including information retrieval, optimization of recommendation engines, video categorization, and video characterization,鈥� Erol said. 鈥淎s part of my Ph.D. research, I am developing another tool, specifically for YouTube video characterization, YouTube video search, and scalable YouTube video categorization to make social computing researchers’ lives easier.鈥� This research is funded in part by the U.S. Air Force Research Lab, U.S. Office of Naval Research, U.S. Army Research Office, U.S. Defense Advanced Research Projects Agency, U.S. National Science Foundation, Arkansas Research Alliance, and the Jerry L. Maulden/Entergy Endowment at the University of Arkansas at Little Rock. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the funding organizations. The researchers gratefully acknowledge the support.]]> /news-archive/2017/09/27/samer-alkhateeb/ Wed, 27 Sep 2017 13:04:03 +0000 /news/?p=68035 ... 糖心Vlog传媒 Little Rock doctoral student presents research on social cyber forensics]]> Samer Al-Khateeb, computer and information sciences doctoral student, has been working as a research and teaching assistant with Dr. Nitin Agarwal, the 糖心Vlog传媒 Little Rock Jerry L. Maulden-Entergy endowed chair and a professor of information science, since 2013 as part of Agarwal鈥檚 research group, 糖心Vlog传媒 Little Rock (COSMOS). 鈥淪amer started working with me in 2013 on U.S. Department of Defense funded projects to investigate how violent, extremist groups, terrorist organizations, state and non-state actors use social media technologies to manipulate public discourse, weaponize narratives, disseminate propaganda, and recruit and radicalize,鈥� Agarwal said. 鈥淗is work on these projects is extremely critical and timely. His research cuts across the disciplinary silos and pushes the fringes of our understanding of emergent behaviors, decentralized and unorganized organizations.鈥� On Sept. 1, Al-Khateeb presented a paper, 鈥淲hen Citizen Journalism Goes Rogue,鈥� about bad citizen journalism cases at the Centre for Media and Celebrity Studies 5th International Conference in New York City. 鈥淭his presentation is about the rise of citizen journalism in the social media era,鈥� Al-Khateeb said. 鈥淭here are many cases where people report events before the traditional media. In many cases, this is a good thing. However, people can also report fake news and spread misinformation. If that news is fake, it can be bad for society. We presented case studies were citizen journalism went rogue.鈥� In July, Al-Khateeb presented his research entitled, “A Social-Cyber Forensics Approach to Understand Twitter & Blogs’ Influence on Anti-NATO Propaganda Campaigns,” during the Social Computing, Behavioral-Cultural Modeling and Prediction Conference held July 5-8 in Washington, D.C. The presentation explored how an online deviant group used different social media channels and techniques to spread propaganda and misinformation against the Trident Juncture Exercise conducted by NATO forces in Europe in 2015. Al-Khateeb used social network measures and social cyber forensics to identify influential information actors and uncover hidden relations between anti-NATO operatives. 鈥淭hese groups use Twitter to build an audience (followers), then they steer their followers to blogs where they frame their narratives,鈥� Al-Khateeb said. 鈥淚n this work, we tried to find the relationships between the online deviant groups on Twitter and their blogs and websites. We did this by running social network analysis and social cyber forensics techniques, which also helped us finding other people who are working with these groups to disseminate propaganda.鈥� The information can help authorities and decision makers see how online groups are coordinating the spread of information and come up with countermeasures and counter narratives to combat these groups. Al-Khateeb and fellow doctoral student Muhammad Nihal Hussain also led a three-hour tutorial for conference presenters on social cyber forensics. At another conference in Washington, D.C., Al-Khateeb presented his paper, “Leveraging Social Network Analysis & Social Cyber Forensics Approaches to Study Cyber Propaganda Campaigns” at the first North American Social Networks Conference of the International Network For Social Network Analysis held July 26-30. Agarwal and Hussain co-authored the paper. Al-Khateeb, who also holds a bachelor鈥檚 degree in computer science and a master鈥檚 degree in applied science from 糖心Vlog传媒 Little Rock, is expected to graduate with his Ph.D. at the end of the semester and has plans to work at a university as a professor and researcher. He has signed a contract with Springer to write a book on social cyber forensics. His research areas include social network analysis, deviant behavioral modeling, deviant cyber flash mobs, cyber propaganda campaigns, social cyber forensics, social computing, data mining in social media, and the detection of bots. He most recently won awards at the 2017 糖心Vlog传媒 Little Rock Engineering and Information Technology Open House, including the Excellence in Research Award, second place in the Most Innovative Award, and second place in the Societal Impact Award.]]>