/news-archive/tag/machine-learning/ 糖心Vlog传媒 Little Rock Wed, 12 Dec 2018 15:35:15 +0000 en-US hourly 1 https://wordpress.org/?v=6.9.4 /news-archive/2018/12/12/kim-tran-social-networks-ipf/ Wed, 12 Dec 2018 15:35:15 +0000 /news/?p=72926 ... 糖心Vlog传媒 Little Rock grad researches validity of patient data found on social networks]]> A University of Arkansas at Little Rock student who is graduating on Dec. 15 is making it easier for medical researchers to validate patient data found on social networks that can be used to make important decisions about what medical products are advanced for future development. When Kim Tran of Little Rock was working at Arkansas Capital Corporation a decade ago, she noted that there was a statewide discussion on the critical importance of computer and information technology. 鈥淚 was working with business and government leaders throughout Arkansas who were talking about technology and how important it was to have access to the infrastructure in order to enable that technology,鈥� she said. 鈥淎t the time, people were also starting to talk about this thing called big data. With this in mind, I wanted to learn more. 糖心Vlog传媒 Little Rock had also just partnered with MIT to develop a curriculum that was focused on the science of data and that is what brought me to 糖心Vlog传媒 Little Rock.鈥� Tran, who began the Ph.D. program in computer and information science in 2010 as a part-time student who worked full time, said one of the most challenging aspects of the process was selecting a topic for her dissertation, citing the more than 1,200 articles she reviewed before choosing a topic. She鈥檚 grateful for her professors who served as mentors during her time at 糖心Vlog传媒 Little Rock. 鈥淒r. Rolf Wigand was always pushing the boundary for me,鈥� she said. 鈥淓very time I felt good about where I was at, he would challenge me to look around the next corner. Ph.D. students need this kind of feedback in order to strengthen the quality of their research. Dr. John Talburt and Dr. Meredith Zozus especially helped me contextualize my research. I also developed lifetime friendships with many professors at the university. They were an exceptionally supportive group, and I was lucky to have that.鈥� Having a support structure during her doctoral endeavors was something she especially owes to her dissertation advisor, Dr. Nitin Agarwal, Maulden-Entergy Endowed Chair and distinguished professor in the Department of Information Science and director of the Collaboratorium of Social Media and Online Behavioral Studies (). 鈥淭he great thing about the Ph.D. process is that you have an advisor who will guide you through the process and help open doors so that you can grow and develop. Dr. Agarwal guided me through the process,鈥� she said. Tran鈥檚 research brings together the fields of machine learning and natural language processing, psychometrics, and social networks, all of which are applied to Idiopathic Pulmonary Fibrosis (IPF), a lung disease which results in scarring (fibrosis) of the lungs for unknown reasons. An estimated five million patients worldwide and 150,000 patients in the United States are affected by this disease. 鈥淜im鈥檚 research bridges the disciplines of statistics, health sciences, information sciences, and social networks by developing a computational framework to assess social media鈥檚 validity in capturing patient reported outcomes from Idiopathic Pulmonary Fibrosis patients,鈥� Agarwal said. 鈥淗er research has far-reaching implications to the health domain by facilitating exploratory efforts in the medical product development process.鈥� Since 2009, regulatory reviewers have been looking at ways to incorporate patient input into its drug selection process, in order to bring drugs to the market sooner, Tran said. In 2015, a discussion held between regulatory reviewers, pharmaceutical companies, and a patient group generated consensus on the potential of social networks in supporting the validity of patient outcomes identified for medical product development and her dissertation creates a scalable framework from which the validity of social networks can be determined. 鈥淗ealthcare is very unique domain since research in this area affects the lives of patients,鈥� she said. 鈥淪o any data you are deriving from any source will require a high level of scrutinization. Social networks are one possible platform that can be used as a source to develop patient-reported outcomes. While the ideal source of feedback is obtained directly from the patient, the way in which this information is gathered is highly variant in scope and in quality. The FDA, for example, still collects patient input through town halls. In the search for more efficient methods of gathering patient understanding, social networks serve as a unique source of observational data.鈥� In order to study whether the data is valid, she uses advanced probabilistic methods to analyze and evaluate Idiopathic Pulmonary Fibrosis messages from Twitter for the last 10 years across 34 different languages from around the world. Tran was drawn to study Idiopathic Pulmonary Fibrosis after attending an international research conference where she spoke with patients and about this little known disease. 鈥淚PF is not as well known or studied as breast cancer,鈥� she said. 鈥淲hen something is idiopathic, you don鈥檛 know the origin. The one thing you do know is that your disease is fatal and that it will result in markedly reduced lung capacity over time. I met with and spoke at length with many patients who were affected by this disease at an international conference. It was eye opening and also touching how driven and motivated these patients were to learn about IPF. They were there because they didn鈥檛 want to just be a patient, they wanted to be a part of finding a cure. That gave me the drive to learn more about the field and to help advance the understanding of idiopathic pulmonary fibrosis.鈥� As part of her dissertation, Tran has collaborate with a researchers across the country who are planning to set up additional studies based on this research. 鈥淚 have been fortunate to meet researchers from other institutions that I have been working with as well as others that I will begin to work with,鈥� she said. 鈥淭his is a group which will bring diverse perspectives and includes researchers from 糖心Vlog传媒MS, Yale, Georgetown, Northeastern, and Tulane University. There is much opportunity to extend this research to fully evaluate the validity of social networks, and I am really looking forward to it.鈥� In the end, Tran is grateful for the opportunities that earning a Ph.D. brought her. 鈥淭he Ph.D. process is an excellent development opportunity as long as you are able to commit to the process,鈥� Tran said. 鈥淭hrough this process, you learn how to learn. I had an opportunity to work across a variety of fields that are all on the cutting edge of things that matter in today鈥檚 business environment and to make a novel contribution to the field.鈥�]]> /news-archive/2018/11/07/google-artificial-intelligence-grant/ Wed, 07 Nov 2018 15:59:22 +0000 /news/?p=72581 ... 糖心Vlog传媒 Little Rock researchers receive 20k grant from Google to study artificial intelligence]]> A University of Arkansas at Little Rock professor and doctoral student have received a $20,000 research grant for their research in artificial intelligence and machine learning.聽 Dr. Xiaowei Xu, professor of information science, and Tolgahan Cakaloglu, a doctoral candidate in computer science, received the grant for their project, “Contextual Advanced Text Representation via Improved Deep Language Model by Utilizing Side Information.鈥� This advanced artificial intelligence technique can be successfully applied to many areas, including natural language processing, information retrieval, and search engines. Xu and Cakaloglu are currently researching how to improve a computer鈥檚 understanding of natural language to improve people鈥檚 experiences with search engines. 鈥淭his technology gives people the opportunity to talk with computers, so the computer can actually answer their questions,鈥� Xu said. 鈥淩ight now, you type in keywords to try to find the answers to your questions. With this model, you can type in questions using plain language. You communicate in a very natural way with the computer instead of trying to complicate things by trying to come up with the right keywords.鈥� The grant allows Xu and Cakaloglu the opportunity to travel to Google鈥檚 headquarters in California to work with the tech giant鈥檚 experts in artificial intelligence research and to conduct experiments using Google Cloud’s TPU/GPU high performance computing clusters, which can run mathematical models with millions of parameters. 鈥淲ithout the Google Grant, for us, it鈥檚 mission impossible,鈥� Xu said. 鈥淲e cannot prove our mathematical computations. These computers will allow us to get real results and show that our program outperforms other programs. I enjoy the collaboration with Google because we get to work on a solution to a real world problem. Industry gives us better and more practical problems to solve.鈥� The researchers see this technology having useful applications across many areas, including customer service, computer-based employee training, and education. 鈥淭here may come a day when people don鈥檛 need to hire a human tutor,鈥� Xu said. 鈥淧erhaps if people have questions about mathematics, physics, chemistry, and other subjects, a computer can hear their questions and give a precise answer and people can tutor themselves. This will be especially valuable for K-12 and higher education.鈥� This research will serve as the basis for Cakaloglu鈥檚 dissertation. Cakaloglu plans to graduate in 2019 and hopes to continue working with Google in future research collaborations. 鈥淲orking with the Google researchers is also very important,鈥� he said. 鈥淎s a researcher, although we follow the trends in AI to solve problems, those people are at the center of AI research, so they know what kind of solution the system needs.鈥� In addition to his research, Cakaloglu founded the Arkansas AI Club, which already has nearly 100 members. 鈥淲e want to raise the public awareness about artificial intelligence in Arkansas and promote collaborations with people interested in the area of artificial intelligence,鈥� Cakaloglu said. 鈥淲e believe there are multiple industry problems that can be solved using AI in Arkansas.鈥� The researchers see a need for more artificial intelligence researchers in the state since they see AI as the next big revolution in technology. 鈥淎rtificial intelligence will change the way people do business,鈥� Xu said. 鈥淚t will be accepted that AI will be the next big revolution, the same way that the Internet changed the world. AI is the next technology that can change the way we live. 鈥淚 think it鈥檚 important to increase the learning of AI across the state so that young talent, high school students, and young professionals can get better educational opportunities and collaborations, so they will have a better future and better jobs. It will hopefully also attract some high tech companies and industry to the area. There is a big shortage of talent. The industry will not be successful here if they cannot find enough people with sufficient expertise in artificial intelligence.鈥�]]> /news-archive/2018/10/22/wei-dai-machine-learning/ Mon, 22 Oct 2018 14:22:01 +0000 /news/?p=72420 ... 糖心Vlog传媒 Little Rock researchers studying development of benchmarks to test machine learning]]> Researchers from the University of Arkansas at Little Rock are joining the ranks of technology companies and universities like Google, Intel, Stanford University, and Harvard University to create the next generation benchmark suite for machine learning.聽 Wei 鈥淒avid鈥� Dai, a doctoral candidate in information science, is pursuing dissertation research on evaluating machine learning with imperfect data. Machine learning is a field of computer science that uses statistical techniques to give computer systems the ability to learn from data. In most experiments, machine learning is tested using perfect data. In this instance, Dai and his dissertation advisor, Dr. Daniel Berleant, professor of information science, are corrupting image files to appear more grainy and pixelated. 鈥淗ow can we evaluate machine learning when data quality has problems? In the real world, imperfect datasets are the majority,鈥� Berleant explained. Thanks to funding from the College of Engineering and Information Technology, Dai traveled to Stanford University on July 18 to attend a technical meeting for MLPerf researchers and was able to discuss his research proposal to develop an imperfect dataset benchmark to test machine learning systems. The MLPerf effort represents a diverse group of universities and industry companies, like Google, Intel, Harvard, and Stanford, that has released the eponymous MLPerf, a new benchmarking tool used to measure the speed of machine learning software and hardware. The MLPerf effort aims to build a common set of benchmarks that enables the machine learning field to measure system performance eventually for both training and inference from mobile devices to cloud services. An accepted benchmark suite will benefit researchers, developers, builders of machine learning frameworks, cloud service providers, hardware manufacturers, application providers, and end users. 鈥淭he important thing is that we want to change from the use of perfect data to imperfect datasets instead,鈥� Dai said. For his dissertation research, Dai plans to create the imperfect data sets, test the performance of the algorithms, and come up with solutions if the algorithms do not work. The ultimate goal of the research is to develop a mutated dataset that can be used as a benchmark researchers can apply in other studies, such as self-driving vehicles, computer vision, and face recognition. Dai earned a master鈥檚 degree in information science from 糖心Vlog传媒 Little Rock in 2016. Before moving to the U.S., he worked as a senior data scientist engineer at IBM China and IBM China Lab for seven years. Dai has published academic papers and books in the U.S. and China. In the U.S., he has published four conference papers, with a fifth currently under review, and has also received three awards from the 糖心Vlog传媒 Little Rock Student Research and Creative Works Expo in 2016 and 2017, and a research award from the Donaghey College of Engineering and Information Technology in 2017. In China, he published two database books and five technical papers, released three patents, and twice won the Excellent Employee Award and Outstanding Instructor Award from IBM. In the upper right photo,聽as part of his research on machine learning, Dai has told the computer to corrupt part of this image of his face. Photo by Ben Krain.]]>