Computer Science and Electrical Engineering
University Of Maryland, Baltimore County

M.S. Thesis Defense

Assessing Confidence in Relation Extraction Systems

Lianjie Sun

2:00ppm Thursday, 27 July 2015, ITE 325b, UMBC

In information extraction, a central and challenging task is extraction of relations. Systems that extract relations from text tend to be very productive, so it is important to quantify confidence or certainty in what is extracted. In this thesis we introduce a framework to assess confidence in relation extraction systems. We trained our system using a logistic regression model based on manually tagged sentences from the New York Times Annotated Corpora. Empirical results based on ROC curves show that our system performs better at computing confidence than previous systems such as Reverb. We conclude with a detailed analysis of the features used in our system and explain how these features might be tailored for use in other relation extraction systems.

Committee: Drs. Tim Oates (chair), Charles Nicholas and Matt Schmill

CSEE professor Jian Chen recently received an award from Department of Defense to to develop new techniques to visualize health informatics data. The award will support two UMBC research students for two years and be done in collaboration with Jesus Caban, Gerard Reidy and Joseph Bleiberg from the Walter Reed National Military Medical Center's National Intrepid Center of Excellence.

The research will the support temporal exploration and analysis of traumatic brain injury and post traumatic stress disorder from patient cohorts. The group will design interactive visualization to move beyond using a small subset of data from the wealth and breadth of clinical information to improve diagnostic accuracy.

The project will help clinicians obtain new insights about the underlying conditions of patients, analyze complex hidden clinical patterns, and visually explore the correlations between many assessment techniques and imaging modalities including neuroimaging, neuropsychiatric measures, patient history, demographic information, and clinical tests).

Last week, Rick Forno, CSEE's Cybersecurity Graduate Program Director and Assistant Director of UMBC's Center for Cybersecurity, joined Australian security journalist Pat Gray for a lengthy discussion on the Risky Business podcast about the OPM data breach, local reactions, and possible US cybersecurity responses to this significant and far-reaching security incident.

This talk was accepted for DEFCON 2015 in Las Vegas later this month. However, for those interested here in the UMBC community, John will conduct an informal preview of his talk on Friday 7/17 at 3:00PM in ITE 366 (DREAM Lab).

"Quantum" Classification of Malware
John Seymour, UMBC

3:00pm Friday 17 July 2015, ITE 366

Quantum computation has recently become an important area for security research, with its applications to factoring large numbers and secure communication. In practice, only one company (D-Wave) has claimed to create a quantum computer which can solve relatively hard problems, and that claim has been met with much skepticism. Regardless of whether it is using quantum effects for computation or not, the D-Wave architecture cannot run the standard quantum algorithms, such as Grover’s and Shor’s. The D-Wave architecture is instead purported to be useful for machine learning and for heuristically solving NP-Complete problems.

We'll show why the D-Wave and the machine learning problem for malware classification seem especially suited for each other. We also explain how to translate the classification problem for malicious executables into an optimization problem which a D-Wave machine can solve. Specifically, using a 512-qubit D-Wave Two processor, we show that a minimalist malware classifier, with cross-validation accuracy comparable to standard machine learning algorithms, can be created. However, even such a minimalist classifier incurs a surprising level of overhead.

John Seymour is a Ph.D. student at the University of Maryland, Baltimore County, where he performs research at the intersection of machine learning and information security. He's mostly interested in avoiding and helping others avoid some of the major pitfalls in machine learning, especially in dataset preparation (seriously, do people still use malware datasets from 1998?) In 2014, he completed his Master’s thesis on the subject of quantum computation applied to malware analysis. He currently works at CyberPoint International, a company which performs network and host-based machine learning, located in Baltimore, MD.

CSEE Ph.D. student Kavita Krishnaswamy is interviewed by Dr. Renetta Tull, UMBC Associate Vice Provost for Graduate Student Development and Postdoctoral Affairs. The taped interview includes several graduate students from UMBC, UMBC's Graduate Dean, Dr. Janet G. Rutledge and Kavita’s parents. Kavita discussed her research and answered questions from the audience regarding graduate school tips, information about research by and for people with disabilities, and motivation to work toward the goal of the PhD. A transcript is available.

Nielsen Audio, a consumer research company that collects and analyzes listener data on radio broadcasting audiences, invites UMBC faculty and students to attend events focused on data science from 11:00am-2:30pm on Thursday June 25 at its headquarters in Columbia MD.

In the past few years, data science has become one of the top career opportunities for students with a background in computing or mathematics, offering interesting challenges and top salaries. Nielsen has been actively recruiting on campus and has hired three graduating UMBC students into its leadership rotational program, as well as several summer interns. They will be recruiting for full-time positions in the Fall.

The Nielsen Data Science Day event will take place in lobby and auditorium of Nielsen Audio's headquarters at 9705 Patuxent Woods Dr #200, Columbia, MD 21046 (map). Activities will include presentations, data science themed games and group discussions.

Between 11:00 and 12:00 participants can engage with interactive games with a Math/Data Science/Audio theme, including Data Science Jeopardy, Name that Tune, and Sampling Marbles. In the auditorium, a short video on data science produced by Nielsen will play continuously.

Lunch is available at Noon, followed by an introduction to data science at Nielsen Audio and presentations from managers of Nielsen's data science groups.

In the afternoon there will be a chance to meet with data scientists and find out what they do and opportunities for internships and positions.

If you have questions, contact the Columbia Data Science Day Committee leads: Kelly Dixon (Sorry, you need javascript to view this email address. ) or Freddie Navarro (Sorry, you need javascript to view this email address. )

Nielsen Audio, a consumer research company that collects and analyzes listener data on radio broadcasting audiences, invites UMBC faculty and students to attend events focused on data science from 11:00am-2:30pm on Thursday June 25 at its headquarters in Columbia MD.

In the past few years, data science has become one of the top career opportunities for students with a background in computing or mathematics, offering interesting challenges and top salaries. Nielsen has been actively recruiting on campus and has hired three graduating UMBC students into its leadership rotational program, as well as several summer interns. They will be recruiting for full-time positions in the Fall.

The Nielsen Data Science Day event will take place in lobby and auditorium of Nielsen Audio's headquarters at 9705 Patuxent Woods Dr #200, Columbia, MD 21046 (map). Activities will include presentations, data science themed games and group discussions.

Between 11:00 and 12:00 participants can engage with interactive games with a Math/Data Science/Audio theme, including Data Science Jeopardy, Name that Tune, and Sampling Marbles. In the auditorium, a short video on data science produced by Nielsen will play continuously.

Lunch is available at Noon, followed by an introduction to data science at Nielsen Audio and presentations from managers of Nielsen's data science groups.

In the afternoon there will be a chance to meet with data scientists and find out what they do and opportunities for internships and positions.

If you have questions, contact the Columbia Data Science Day Committee leads: Kelly Dixon (Sorry, you need javascript to view this email address. ) or Freddie Navarro (Sorry, you need javascript to view this email address. )

Nielsen Audio, a consumer research company that collects and analyzes listener data on radio broadcasting audiences, invites UMBC faculty and students to attend events focused on data science from 11:00am-2:30pm on Thursday June 25 at its headquarters in Columbia MD.

In the past few years, data science has become one of the top career opportunities for students with a background in computing or mathematics, offering interesting challenges and top salaries. Nielsen has been actively recruiting on campus and has hired three graduating UMBC students into its leadership rotational program, as well as several summer interns. They will be recruiting for full-time positions in the Fall.

The Nielsen Data Science Day event will take place in lobby and auditorium of Nielsen Audio's headquarters at 9705 Patuxent Woods Dr #200, Columbia, MD 21046 (map). Activities will include presentations, data science themed games and group discussions.

Between 11:00 and 12:00 participants can engage with interactive games with a Math/Data Science/Audio theme, including Data Science Jeopardy, Name that Tune, and Sampling Marbles. In the auditorium, a short video on data science produced by Nielsen will play continuously.

Lunch is available at Noon, followed by an introduction to data science at Nielsen Audio and presentations from managers of Nielsen's data science groups.

In the afternoon there will be a chance to meet with data scientists and find out what they do and opportunities for internships and positions.

If you have questions, contact the Columbia Data Science Day Committee leads: Kelly Dixon (Sorry, you need javascript to view this email address. ) or Freddie Navarro (Sorry, you need javascript to view this email address. )

Military's Use Of Social Security Numbers A Vulnerability For Veterans
By Patricia Murphy – KUOW (Seattle Public Radio)

A massive data breach at the federal Office of Personnel Management has exposed the Social Security numbers and personnel records of nearly every federal worker. The implications for federal employees, military service members and the intelligence community could be extraordinary.  

But at a very basic level U.S. service members have been at high risk for identity theft for decades.

In the military your nine-digit service ID number is used for everything, whether you're checking out gym equipment or picking up your laundry. Since the late 1960s that number has been a service member’s Social Security number.

Rick Forno is the assistant director of the University of Maryland Baltimore County Center for Cybersecurity. He said using Social Security numbers as a unique identifier isn't new….

Source: KUOW

Hiring Students for Summer Text Mining Project

A new, interdisciplinary research project offers a total of five positions to graduate or undergraduate students from multiple disciplines who exhibit the right combination of initiative, skills and reliability. The project investigates undergraduate teaching at UMBC, and is led by the combined expertise of the Shriver Center, Interdisciplinary Studies Program and the Honors College. The students will work as an integrated, interdisciplinary team, and may expect to expand their own knowledge, skills and experience as a result. UMBC graduate students and exceptionally qualified undergraduate students will be considered.

A group of Academic and Student Affairs Division faculty and staff are working to better track, assess, strengthen and increase, and recognize and reward applied learning experiences across the UMBC Campus. In FY15, this group’s work has focused on creating a plan for assessing the impact of applied learning experiences (broadly defined) on students’ affective development.

Each position offers partial summer support funded by a Hrabowski Fund for Innovation: Implementation and Research grant.

80 hours (may span across 4-8 weeks depending on availability-no more than 20 hrs/week)

Stipend of $1150

The five graduate students we seek will have a background in the following skills/disciplines: Information Systems; Computer Science; Human Centered Computing; Instructional Systems Design; Applied Mathematics; Statistics; Human Services Psychology; Applied Sociology; Education; Language, Literacy & Culture; and Public Policy.

The project aims to continue for a minimum of three years, expanding subject to further successful funding. Students who perform well may be invited to extend their role.

One goal is to explore, through written student feedback, how diverse courses have contributed to their perception of growth in key attributes.  This first summer of the project will look at pilot data, exploratory analysis of scrappy, inconsistent, self-reported text, in order to figure out how to collect better, more uniform data in the Fall semester. One of the skill sets we wish for in the team would be a computer scientist able to mine text to seek the suggestion of themes, patterns or clues to what we would need to ask to get such patterns more clearly in the future.

All interested candidates should send a letter of interest, resume with references, and a statement with general summer weekly availability (e.g., Monday, Wednesdays & Fridays from 9am – 2pm) to Michele Wolff at Sorry, you need javascript to view this email address. by Wednesday, June 24.  Exceptionally qualified undergraduate students will also be considered.