Northeastern University College of Computer and Information Science Thu, 17 Apr 2014 18:40:19 +0000 en-US hourly 1 ‘Huntington 100’ Students Honored for Their Achievements Mon, 14 Apr 2014 13:36:08 +0000 040914_BC_Huntington100_044Northeastern’s newest “Hunt­ington 100,” a group of extra­or­di­nary stu­dents selected for their impres­sive achieve­ments and impact both on campus and around the world, was hon­ored last week at a recep­tion with uni­ver­sity leaders, fac­ulty, and staff.

The honor, now in its second year, rec­og­nizes seniors and under­classmen who have excelled in their respec­tive areas—from research and entre­pre­neur­ship to expe­ri­en­tial learning and ath­letics. This year’s newest “100” mem­bers were joined at the recep­tion by under­classmen who received the dis­tinc­tion last year.

We’re very proud of your accom­plish­ments,” Aoun said. “You were chosen because you are role models, and you’ve had a great impact on the uni­ver­sity, stu­dents, fac­ulty, staff, and society.”

Many grad­u­ating seniors are set to begin exciting careers or pres­ti­gious grad­uate pro­grams. Aoun urged soon-​​to-​​be-​​graduates to remain con­nected with North­eastern as alumni and net­work with fellow alums across the globe. “Your job is just begin­ning,” he said. “No matter where you go, you will be a North­eastern grad­uate and a leader. Take a piece of North­eastern with you.”

Lead­er­ship was a common char­ac­ter­istic among the group, which included Stu­dent Gov­ern­ment Asso­ci­a­tion Pres­i­dent Nick Naraghi, CIS’15; women’s soccer cap­tain Hanna Terry, SSH’14; Max Kaye, DMSB’14, CEO of IDEA, Northeastern’s student-​​run ven­ture accel­er­ator; Stanislas Phanord, SSH’14, who recently received the esteemed Rangel Fel­low­ship and Ful­bright Schol­ar­ship; and Laura Marelic, AMD’15, who has spear­headed the cre­ation of the new stu­dent design agency SCOUT.

Many “Hunt­ington 100” stu­dents have also had dynamic co-​​op expe­ri­ences. Nate Bessa, CIS’14, devel­oped a soft­ware pro­gram that mon­i­tors a physi­cian pro­duc­tivity incen­tive pro­gram at Brigham and Women’s Hos­pital. Rachael Tompa, E’14, worked at NASA’s Jet Propul­sion Lab­o­ra­tory on the Thermal Tech­nology and Fluid Sys­tems group. Klevis Xharda and Laura Mueller-​​Soppart, both SSH’14, were selected to com­plete expe­ri­en­tial learning oppor­tu­ni­ties at the White House.

Stu­dents hailed co-​​op as both a pri­mary factor in their choosing North­eastern and their pro­fes­sional growth during their time here. Keith Ray­burn, E’14, com­pleted three co-​​ops at CDM Smith, a Cam­bridge, Mass.-based con­sulting, engi­neering, con­struc­tion, and oper­a­tions firm. On his third co-​​op, he was tasked with leading a sewer system reha­bil­i­ta­tion project.

In addi­tion to his work on co-​​op, he’s a member of Northeastern’s chapter of Engi­neers Without Bor­ders, which has brought clean water to fam­i­lies in Hon­duras and Bbanda, Uganda since its founding in 2005. This spring, he accepted a full-​​time posi­tion as an envi­ron­mental engi­neer at Envi­ron­mental Part­ners Group in Quincy, Mass. “The on-​​the-​​job training was so valu­able for me, just learning the day-​​to-​​day oper­a­tions,” he said.

Michele Bellini, DMSB’14, was part of the stu­dent lead­er­ship team that orga­nized the university’s inau­gural Global Summit on the 2008 finan­cial crisis held on campus ear­lier this month. The Italy-​​native is also in the D’Amore-McKim School of Busi­ness’ BSIB pro­gram and chapter of Net Impact, an inter­na­tional non­profit orga­ni­za­tion whose mis­sion is to make a pos­i­tive impact on society by growing and strength­ening a com­mu­nity of leaders who use busi­ness to improve the world.

I’ve expanded my hori­zons here,” he said. “North­eastern really works hard to empower its students.”

]]> 0
New Experiential Master’s Program Aligns Talent with High-​​Demand Industries Mon, 14 Apr 2014 13:29:29 +0000

Creative IndustriesNorth­eastern Uni­ver­sity has launched a new master’s degree pro­gram with a built-​​in expe­ri­en­tial learning com­po­nent, specif­i­cally designed for pro­fes­sionals who want to tran­si­tion into high-​​growth indus­tries, but do not have under­grad­uate degrees or work expe­ri­ence to match the needs of these industries. The pro­gram, called ALIGN (Accel­er­ated Link to Industry through Northeastern’s Global Net­work), offers eight master’s degrees in cutting-​​edge indus­tries, such as bioin­for­matics, health infor­matics, and cyber­se­cu­rity. Each master’s degree has a cus­tomized cur­riculum that bridges the student’s under­grad­uate edu­ca­tion with the grad­uate degree con­tent and includes a six to nine month co-​​op posi­tion with one of Northeastern’s global net­work of 3,000 employers.

The ALIGN pro­gram is the only expe­ri­en­tial grad­uate degree pro­gram in the U.S. designed for stu­dents with under­grad­uate degrees in a variety of dis­ci­plines who want to switch careers and need both an advanced degree and the real-​​world expe­ri­ence to do so. The pro­gram lever­ages Northeastern’s lead­er­ship in com­bining real-​​world pro­fes­sional expe­ri­ence with rig­orous class­room learning and uti­lizes the university’s global co-​​op infra­struc­ture and three domestic campuses.

Deliv­ered in a hybrid format at Northeastern’s Boston, Char­lotte, N.C., and Seattle, Wash. cam­puses, the eight master’s degrees include bioin­for­matics, com­puter sci­ence, energy sys­tems, engi­neering man­age­ment, health infor­matics, infor­ma­tion assur­ance, project man­age­ment, and reg­u­la­tory affairs for drugs, bio­logics, and med­ical devices—fields that closely align Northeastern’s aca­d­emic and research strengths with the talent needs of high-​​demand global industries.

The inte­gra­tion of advanced learning and real-​​world expe­ri­ence is a pow­erful com­bi­na­tion,” said Joseph E. Aoun, pres­i­dent of North­eastern Uni­ver­sity. “Northeastern’s net­work of the most inno­v­a­tive industry part­ners world­wide puts our grad­uate stu­dents in an unmatched posi­tion to lead in a global economy.”

The ALIGN pro­grams have been strate­gi­cally selected to match key indus­tries with excep­tional employ­ment oppor­tu­ni­ties, long-​​term career poten­tial, and cutting-​​edge chal­lenges. For example, the global bioin­for­matics market is pro­jected to nearly quadruple in the next four years.

North­eastern piloted the ALIGN pro­gram in its Seattle, Wash. campus with a master of sci­ence in com­puter science—a crit­ical demand in sev­eral high-​​growth indus­tries. The degree is designed to build on stu­dents’ strong quan­ti­ta­tive and ana­lyt­ical skills with intro­duc­tory courses in pro­gram design and com­puter sys­tems. While com­puter sci­ence grad­uate stu­dents typ­i­cally grad­uate from pro­grams focused on math and engi­neering, Northeastern’s pro­gram is designed for stu­dents from a broad range of under­grad­uate majors, including his­tory, eco­nomics, and physics.

ALIGN builds on Northeastern’s lead­er­ship in global expe­ri­en­tial learning and enables par­tic­i­pating stu­dents to take advan­tage of the university’s net­work of more than 200,000 alumni and 3,000 co-​​op employers.


]]> 0
Video: PhD Program Energy 24/7 Thu, 27 Mar 2014 13:25:48 +0000 The Ph.D. program at the College of Computer and Information Science has a rapidly growing faculty engaging in cutting edge research with our students. Watch the video below to see how our program truly has Energy 24/7:

]]> 0
Beware this Big iOS flaw — and it’s Not Alone Mon, 17 Mar 2014 13:32:01 +0000 A change that Apple imposed to make iOS 7 more secure instead has dramatically weakened the security of devices running that mobile operating system, a security researcher has charged.

At the CanSecWest conference here last week, Azimuth Security researcher Tarjei Mandt said that Apple made a major mistake when it changed its random-number generator to make its kernel encryption tougher in iOS 7. The kernel is the most basic level of an operating system and controls things like security, file management, and resource allocation.

“In terms of security, it’s much worse than iOS 6,” Mandt said. Soon after his presentation Wednesday in the Grand Ballroom of the Sheraton Wall Centre, he published his presentation slides (PDF) and supporting whitepaper (PDF) as evidence.

And in a testament to the enduring challenges of getting mobile security right, other presentations at CanSecWest also called attention to flaws in the Android and in BlackBerry 10 mobile OSes.

 How Apple left the iOS 7 kernel vulnerable

The technical and complicated change boils down to how Apple calculates randomly generated numbers used in the encryption of the kernel. If the numbers can be guessed, their randomness is irrelevant, and the kernel — key to control of the computer, or in this case the iOS phone or tablet — can be compromised.

Apple, he explained, recognized that the method of generating random numbers in iOS 6 could be improved on. Its security engineers leveraged the phone’s CPU clock counter on earlier version of iOS, Mandt said.

“That’s not very good, but still somewhat unpredictable,” he said.

The problem with the new generator in iOS 7 is that it uses a linear recursion algorithm, Mandt said, which has “more correlation” between the values it generates. That makes them easier to extrapolate and guess, he said.

“Normally, you shouldn’t be able to get any of these values in the first place,” Mandt said.

The kernel exploit is severe, although Mandt did not pair it with a vulnerability. Still, that means that anybody who can find an unpatched vulnerability in iOS 7, such as the “goto fail” vulnerability that was patched last month, can gain kernel-level access.

Apple appears to be taking the flaw seriously, but did not return a request for comment. CNET will update the story when the company responds.

“Apple [security engineers attending CanSecWest] approached me afterwards and they appeared to be kind of concerned,” he said. But he cautioned that this exploit should not be underestimated, and that left unfixed, it would effectively roll back 10 years of security-hardening techniques in iOS.

Using jiu-jitsu to fix Android fragmentation flaws

An Android presentation just after Mandt’s asserted that the one-two punch of Android fragmentation has placed Android users at risk of missing out on important security updates. That’s not going to be fixed anytime soon, they said.

The issue, argued Jon Oberheide of Duo Security and Northeastern University security researcher Collin Mulliner, lies in how Android devices receive — or more precisely, don’t receive — their updates.

 ”The Chrome guys will deliver an update within 24 hours. On Android, it can take months and years,” said Oberheide. “Your carrier doesn’t have a lot of incentive to fix your ancient HTC Evo. They want you to buy the latest and greatest device.”

So, the pair said, even when Google patches Android security flaws, the handset manufacturer and the carrier effectively stop patches from reaching the people who need them.

Android security apps can’t be relied on, Mulliner said, because they’re fighting Android malware — something that he said just isn’t a big problem in most regions.

“None of the big antivirus or security companies are doing a really good job because they’re all concerned with stopping malware,” he said.

Another dead-end, he said, is that Android architecture “doesn’t allow” partial updates.

“Google should be able to update anything that’s not kernel, but to do that you have to separate everything much better in the code,” Mulliner said. “Technically, it’s possible, but I could see the manufacturers not wanting to allow that because then you lose part of the device.”

However, they did hit on a method that flips unpatched exploits into tools for patching the bugs. Starting work at the end of 2012, it uses third-party vulnerability patches, independent of both device and Android version.

“Version numbers don’t tell you anything anymore, whether you buy one device for yourself or 100 devices for your company,” Mulliner said.

“There’s one patch for many devices, with no performance problems, and the patch self-contained,” Oberheide told the crowd.

Their first app, called ReKey, delivers a fix for the MasterKey bug.

They’ve built it to require the owner to root the phone first, so that it can’t be turned into a universal malware delivery tool, and they caution that it’s not for all Android owners. People who use Nexus devices and third-party custom ROMs such as CyanogenMod generally get updates much faster than the rest of Android owners.

Afterward, Mulliner dispensed some advice to people who want to buy Android phones. From a security point of view, he said, “Buy only Nexus devices.”

New BlackBerry era brings new risks

The QNX-based BlackBerry 10 is a major change for BlackBerry for many reasons, not the least of which are its security implications. The hardened, security-focused platform of the legacy BlackBerry OS made that the ideal mobile operating system for large businesses and governments.

Security researchers Ben Nell of Accuvant and Zach Lanier of Duo Security said that BlackBerry 10 opens itself up in ways that previous versions of BlackBerry didn’t, because it was fused on top of QNX, which powers everything from space shuttles to car operating systems.

“Some of the security enhancements introduced in BlackBerry 10 might help mitigate core issues in QNX, but not in other iterations of QNX,” said Lanier. But combining QNX with legacy BlackBerry, he said, “they inherited some bugs, fixed others, and introduced some new ones.”

Lanier noted that BlackBerry is pushing for QNX to be the top “Internet of Things” platform, but by combining QNX with mobile BlackBerry, the company could be putting at risk some of the infrastructure implementations of QNX — such as power plants.

“You may not want to shut down a nuclear reactor [running QNX] for maintenance,” he said. “If QNX continues to gain popularity outside of BlackBerry 10, there will be other issues that will crop up.”

One area where BlackBerry 10 is at risk, they said, is app permissions. “There are a couple of permissions that are in there that allow sockets to be open to the Internet. Any app can do it,” they said.

“We did report things like oversights in file permissions,” said Nell, “the sorts of things that were related to legacy bugs.”

Another problem with BlackBerry 10 is that minor vulnerabilities and weaknesses, along with some of those legacy bugs, could be chained together to cause bigger problems.

“If you had corporate email on the device, we could read your corporate email. It was a series of small issues that chained together were a bigger problem,” Nell said. He was reluctant to talk further about specifics, for fear of revealing problems that haven’t yet been reported publicly.

Nell got his start on BlackBerry research when his company was hired by a potential BlackBerry client to check out how secure BlackBerry 10 actually was. He and Lanier refused to discuss specifics because of a two-year nondisclosure agreement that Nell signed, a common practice in the security world.

Other areas they thought of looking at, but wouldn’t comment on because of their nondisclosure agreement, included checking on how the processes communicate with each other, and privilege escalations of the kind that had Mandt looking into iOS. A bug of either of those types likely would apply to both BlackBerry and QNX.

In the end, Lanier said that he’d recommend BlackBerry 10 as an enterprise device, but not for “bring your own device” customers. Unfortunately for BlackBerry, the dual-use feature of BlackBerry 10 that lets owners switch between work and home modes was designed to appeal precisely to the BYOD crowd.

“To BlackBerry’s credit,” said Lanier, “they baked in all the management features, the separation of work data and user data, from the get go. Now if only they could get people to use it.”

Article from

]]> 0
Does Big Data Have the Flu? Fri, 14 Mar 2014 13:13:27 +0000
bigdata2These days, when people start feeling a fever and a sore throat coming on, often times their first move isn’t to the med­i­cine cab­inet. Instead, it’s to a com­puter or smart­phone to Google their symptoms.

These queries, which make up only a tiny frac­tion of the more than 7 bil­lion total queries the search engine han­dles each day, are all stored by Google. The com­pany uses this data for a variety of rea­sons; it can help Google improve its search results for users—which also boosts the company’s bottom line—and can also ben­efit the pop­u­la­tion as a whole in other ways.

One example of the latter is Google Flu Trends, a sta­tis­tical model devel­oped by engi­neers at—the company’s foun­da­tional arm—in an effort to “now-​​cast” what’s hap­pening with the flu on any given day.

But research has shown that GFT often misses its target. These results led North­eastern Uni­ver­sity net­work sci­en­tists and their col­leagues to take a closer look at how Big Data should be used to advance sci­en­tific research. Their report was pub­lished online Thursday in the journal Sci­ence.

“Big Data have enor­mous sci­en­tific pos­si­bil­i­ties,” said North­eastern pro­fessor David Lazer. “But we have to be aware that most Big Data aren’t designed for sci­en­tific pur­poses.” Fully achieving Big Data’s enthu­si­as­ti­cally lauded poten­tial, he added, requires a syn­thesis of both com­puter sci­ence approaches to data as well as tra­di­tional approaches from the social sciences.

The paper was co-​​authored by Lazer, who holds joint appoint­ments in the Depart­ment of Polit­ical Sci­ence and the Col­lege of Com­puter and Infor­ma­tion Sci­ence; Alessandro Vespig­nani, the Stern­berg Family Dis­tin­guished Uni­ver­sity Pro­fessor of Physics at North­eastern who has joint appoint­ments in the Col­lege of Sci­ence, Bouvé Col­lege of Health Sci­ences, and the Col­lege of Com­puter and Infor­ma­tion Sci­enc; North­eastern vis­iting research pro­fessor of polit­ical sci­ence Ryan Kennedy; and Gary King, a pro­fessor in the Har­vard Uni­ver­sity Depart­ment of Government.

“In a sense, Google Flu Trends is not bad, but it’s no better than any basic approach to time series pre­dic­tion,” Vespig­nani said. “So the issue is in the claims and the dis­re­gard of other tech­niques or data more than the actual result.”

In their paper, the researchers explain where Google Flu Trends went wrong and examine how the research com­mu­nity can best uti­lize the out­puts of Big Data com­pa­nies as well as how those com­pa­nies should par­tic­i­pate in the research effort.

By incor­po­rating lagged data from the Cen­ters for Dis­ease Con­trol and Pre­ven­tion as well as making a few simple sta­tis­tical tweaks to the model, Lazer said, the GFT engi­neers could have sig­nif­i­cantly improved their results. But in a com­panion report also released Thursday on the Social Sci­ence Research Network—an online repos­i­tory of schol­arly research and related materials—Lazer and his col­leagues show that an updated ver­sion of GFT, which came about in response to a 2013 Nature article revealing GFT’s lim­i­ta­tions, does little better than its predecessor.

While Big Data cer­tainly holds great promise for research, Lazer said, it will only be suc­cessful if the methods and data are made—at least partially—accessible to the com­mu­nity. But that so far has not been the case with Google.

“Google wants to con­tribute to sci­ence but at the same time does not follow sci­en­tific praxis and the prin­ci­ples of repro­ducibility and data avail­ability that are cru­cial for progress,” Vespig­nani said. “In other words they want to con­tribute to sci­ence with a black box, which we cannot fully scru­ti­nize and understand.”

If sci­en­tists are to “stand on the shoul­ders of giants,” as the old adage requires for moving knowl­edge for­ward, they will need some help from the giants, Lazer said. Oth­er­wise fail­ures like that with Google Flu Trends will be ram­pant, with the poten­tial to tar­nish our under­standing of any­thing from stock market trends to the spread of disease.

]]> 0
Who’s Got Your Digital Dossier Wed, 12 Mar 2014 15:17:40 +0000 When users’ per­sonal infor­ma­tion is shared with a web­site or mobile appli­ca­tion, that data enters some­thing of a gray market with many unknowns about where that infor­ma­tion will end up or how it will be used. What’s more, indi­vid­uals have few options for con­trol­ling what is shared with third par­ties, says Dave Choffnes, a mobile sys­tems expert and assis­tant pro­fessor of com­puter and infor­ma­tion sci­ence at North­eastern Uni­ver­sity. Choffnes is devel­oping soft­ware that aims to pro­vide data to researchers hoping to tackle this issue and pro­vide some level of con­trol to users. We asked him about the extent and impli­ca­tions of this third-​​party data sharing.

How are users’ data currently being shared online, and how much are they aware of and able to control?

The short answer is that we don’t know how much sharing is going on. We know that information is collected by businesses, and we know that this information is being monetized through advertising. For example, when I search a travel site for a plane ticket to a destination, I see ads related to that destination on other websites. But surely this kind of sharing is only the tip of the iceberg.

In general, I think users—average or otherwise—are aware of frighteningly little. There are two problems: we lack transparency into what is being shared and have almost no control over how it is shared. Several tools attempt to improve transparency. One, Collusion, allows you to track the trackers for desktop Web browsing. Another, called Meddle, which my team developed, can do the same for mobile app traffic.

We don’t have great tools for controlling what is being shared. There are a number of initiatives to allow consumers to opt out of advertising (e.g., Do Not Track and NAI Opt Out). However, there is no enforcement of such preferences and advertiser participation is voluntary. The problem is one of incentives: the advertisers have little reason to stop tracking if there is no downside for them. Governmental policy changes can help, but we also need better tools to allow users to take control of how they are tracked. With Meddle, we are trying to do the latter.

What are the advantages and disadvantages to having our data shared in this manner?

Data sharing and advertising can generate revenue that allows popular apps and services to be “free” for users—a large fraction of services we enjoy today would not exist without such revenue streams. Of course, this is free as in dollars, but not free as in freedom, since the users relinquish control of potentially sensitive data. The disadvantage is that once data is shared, it is nearly impossible to “unshare” it and we do not yet understand the risks. For example, once shared, this private data may be breached by malicious parties. We have seen negative implications of private data being made public. For example, see this story in Wired.

What needs to be done to improve the state of privacy in this arena, and how much is the user’s responsibility to make that happen?

It seems clear that the current state of affairs is not optimal. As a user, I would like to see more control put into our own hands and I would like it to be free to do so. Realistically, I think users, companies, and policymakers will need to share the responsibility of protecting privacy. This entails revisiting current policies for transparency, control, and pricing when it comes to decisions regarding collecting and using data gathered from users.

I think a key issue is that no one will act to improve privacy until being made aware of potential (and actual) problems that stem from today’s data gathering. To address this, we need to find ways to make such information easily accessible to average users and policymakers. Collusion took a huge leap forward in this direction, and we are trying to build upon this success in Meddle.


]]> 0
New Tools to get Your Smartphone up to Speed Wed, 26 Feb 2014 14:45:34 +0000 David Choffnes

“Check the map,” says the voice in the com­mer­cial. “Verizon’s super fast 4GLTE is the most reli­able and in more places than any other 4G network.”

But is it truly the most depen­dent, the most widespread?

In fact, there’s no way of knowing: “There’s absolutely no useful quan­ti­ta­tive data for com­par­ison pur­poses,” according to mobile sys­tems expert Dave Choffnes, an assis­tant pro­fessor in Northeastern’s Col­lege of Com­puter and Infor­ma­tion Sci­ence. And that means we have no con­trol over the per­for­mance and reli­a­bility of our mobile Internet use.

Last year, the number of people using mobile devices to access the Internet sur­passed the number of people using desktop com­puters to log online. While “most eye­ball time is on devices,” Choffnes said, mobile per­for­mance is nowhere near what we expect from our desktop expe­ri­ences. Researchers like him would need much more data to even begin bringing mobile Internet use up to speed—literally.

“Ide­ally we’d have mea­sure­ments from every­where on every net­work all the time,” he said. “And then we’d want to cor­re­late that with where we use our phones and what appli­ca­tions we use on our phones. And then you can imagine with all that infor­ma­tion we could spit out a number and say this car­rier is going to give you the best overall performance.”

But col­lecting that data is easier said than done—presumably, mobile providers would have done it by now if it weren’t. One hurdle is that users need to opt-​​in to donating their data, explained Choffnes, since doing so means giving up some of their expen­sive data-​​plan as well as pre­cious bat­tery life. As a result, he said, “we have to give an incentive.”

To that end, Choffnes has cre­ated a code that could send data to the devel­opers of a mobile app once it has been installed on a smart­phone. Devel­opers could incor­po­rate this code into what­ever apps they create, but users will only opt in to those that are the most enter­taining and useful.

For instance, Choffnes envi­sions an app that would allow two users to pit their mobile speeds against each other. Or one that pro­vides real-​​time infor­ma­tion about a phone’s per­for­mance to inform which net­work provider one should choose.

In another project, which he’s calling Meddle, Choffnes is using the same approach to incen­tivize users into donating their data for research. Along with lim­ited per­for­mance and reli­a­bility con­trol, mobile users also lack con­trol over how their apps share their data with the rest of the Internet.

Meddle is not itself an app, but rather a proxy net­work through which all of a phone’s traffic must pass before accessing a web­site or app. Tra­di­tion­ally, the pri­vacy (or lack thereof) of that journey is gov­erned by the app through which the user is trav­eling. Meddle encrypts every­thing it sends and receives, which instantly improves a user’s pri­vacy experience.

Addi­tion­ally, Meddle allows users to view their activity—and the activity of their apps—online. When we click “agree” to an app’s terms, we often give it per­mis­sion to share our data with other parts of the web. Meddle users can view a map of that activity and shut down any unwanted data sharing, pro­vided it doesn’t inter­fere with the app’s functionality.

In exchange for these improve­ments, users agree to share all their activity data with Choffnes, but only after it’s been scrubbed for anonymity. Choffnes then uses it to inves­ti­gate the per­for­mance, reli­a­bility, and pri­vacy of mobile net­work traffic. “If we can’t see what devices are doing, we can’t opti­mize them to work well all the time,” he said.


]]> 0
Face Value Tue, 25 Feb 2014 14:21:20 +0000 Raymond FuFace­book is home to nearly 3 bil­lion photos. Every minute, YouTube grows by another 100 hours of video. And, according to IHS Research, some 30 mil­lion sur­veil­lance cam­eras pepper our public spaces, col­lecting nearly 4 bil­lion hours of footage each week. Need­less to say, there’s a lot of image data that’s ripe for the picking.

Con­tent like this helped break crim­inal cases such as the 2013 Boston Marathon bombing. But if we want to carry on with sim­ilar suc­cesses, we’ll need ever more sophis­ti­cated algo­rithms to parse the data deluge.

For his part, North­eastern Uni­ver­sity assis­tant pro­fessor Ray­mond Fu is working to improve the cur­rent state-​​of-​​the-​​art of bio­met­rics soft­ware, which auto­mat­i­cally dis­tin­guishes between dif­ferent cat­e­gories of people as well as between indi­vid­uals themselves.

Fu’s research recently earned him one of two Young Inves­ti­gator awards from the Inter­na­tional Neural Net­work Society in 2014. “This is a real honor and inspires me to keep up the good work,” said Fu, a machine-​​learning expert who holds joint appoint­ments in the Col­lege of Engi­neering and the Col­lege of Com­puter and Infor­ma­tion Sci­ence.

Backed by funding from Sam­sung Research of America, the research and devel­op­ment arm of the inter­na­tional elec­tronics com­pany, Fu has recently begun devel­oping visual recog­ni­tion soft­ware for use on social media net­works such as Face­book and Twitter.

“When people share facial images on social net­works, those images are in the wild. So you have uncon­strained data—meaning it’s not col­lected in the lab under con­trolled con­di­tions,” said Fu. “It can be from mul­tiple cam­eras, mul­tiple resources, so the data has a lot of variables.”

To cir­cum­vent this problem, his algo­rithm ranks the infor­ma­tion in all of the images and quickly tosses out any out­liers. “If some­thing is very dif­ferent from the rest of the images, our algo­rithm can rule it out and mit­i­gate noise,” he explained.

The soft­ware can “learn” a person’s unique face and use that infor­ma­tion to leverage the vast stores of image data online to under­stand society or inform inves­ti­ga­tions. For instance, Fu’s algo­rithms could help iden­tify what types of people turn out at a protest, he said, by rec­og­nizing gen­eral char­ac­ter­is­tics rather than indi­vidual ones: Are the people pho­tographed at demon­stra­tions such as Occupy Wall Street car­rying cam­eras and note­books, and thus likely jour­nal­ists? Are there more uniform-​​donning policemen than protestors?

Of course, adver­tisers and cor­po­ra­tions could also use this data for less-​​noble pur­suits, such as tar­geting their prod­ucts at par­tic­ular groups or indi­vid­uals. “There is always a trade off between pri­vacy and ser­vices,” said Fu. “Every­thing I’m doing uses data that’s pub­li­cally avail­able. We’re trying to pro­vide the best models for ana­lyzing it.”

It’s up to the rest of us—you and me and our representatives—to deter­mine how we should use those models.


]]> 0
Brilliant and Motivated, but a Good Hire? Tue, 25 Feb 2014 14:17:19 +0000 The evi­dence is clear that Albert László Barabási, a world-​​renowned net­work sci­en­tist and Dis­tin­guished Uni­ver­sity Pro­fessor of Physics at North­eastern Uni­ver­sity, has enjoyed a suc­cessful career. As the founding pro­fessor of Northeastern’s net­work sci­ence pro­gram, Barabási is a “bril­liant and moti­vated” scholar, in the words of his grad­uate advisor, Gene Stanley, him­self a dis­tin­guished pro­fessor of physics at Boston University.

Barabási has pub­lished four books and 142 papers, which have col­lec­tively received more than 100 thou­sand cita­tions. He’s even got a Kevin Bacon number of one, thanks to his appear­ance along­side the Hol­ly­wood actor in a movie called “Con­nected,” according to Larry Finkel­stein, dean of the Col­lege of Com­puter and Infor­ma­tion Sci­ence. As Murray Gibson, dean of the Col­lege of Sci­ence, put it, “you can’t hide from the obvious impact of his work.”

But on Monday at a cer­e­mony installing him as the inau­gural Robert Gray Dodge Pro­fessor of Net­work Sci­ence, Barabási asked whether all these acco­lades actu­ally made him a good hire back in 2009.

In his inau­gural lec­ture that focused on the “sci­ence of suc­cess,” Barabási noted the mea­sures are con­sis­tently used to deter­mine a researcher’s suc­cess in the sci­en­tific com­mu­nity: a journal’s impact factor and a paper or researcher’s total cita­tion count. “We love to hate these num­bers,” explained Barabási, because they don’t actu­ally do a very good job of pre­dicting the future impact of a paper or suc­cess of a career.

In a new line of research for his lab, Barabási’s team is devel­oping a more math­e­mat­i­cally robust way of mea­suring suc­cess. They’ve exam­ined it in the con­text of physics research papers and are begin­ning to expand it into other broad-​​ranging areas, including the suc­cess of a Twitter feed.

Thank­fully to those in atten­dance Monday who hired him five years ago, Barabási can use these methods to pre­dict con­tinued suc­cess for his own career.

While a researcher’s single block­buster suc­cess often comes in his or her first 10 years on the job, there is actu­ally no pre­dic­tive value in that time­line for future suc­cess. Based on pro­duc­tivity, he said, “we can’t see suc­cess coming nor do we really learn from it.”

Instead, things like an individual’s “excel­lence parameter”—a mea­sure of how he com­pares to his peers in the way he addresses the world’s challenges—is much more impor­tant than their total cita­tions, the impact of the jour­nals they pub­lish in, or even their overall pro­duc­tivity. We can find an accu­rate mea­sure of suc­cess, Barabási said, “you just have to look at the right variables.”

In closing remarks, Stephen W. Director, provost and senior vice pres­i­dent for aca­d­emic affairs, noted that Robert Gray Dodge, for whom Barabási’s new pro­fes­sor­ship is named, was essen­tially the university’s first pro­fessor. “I think it’s fit­ting that he started some­thing brand new and achieved great suc­cess,” said Director, noting that is pre­cisely what Barabási has done since joining the North­eastern fac­ulty and spear­heading the nation’s first pro­gram in net­work science.

“One thing I learned through net­work sci­ence,” Barabási said, “is that a net­work is use­less as a set of nodes. It’s all about the links.” North­eastern, he said, has allowed him to pursue those links in order to main­tain the high level of work in which his lab is engaged.

Pres­i­dent Joseph E. Aoun noted that before com­mencing his tenure at North­eastern, Barabási demanded only one thing: “an envi­ron­ment con­ducive to building the best pro­gram in the country.” When Aoun asked how he pro­posed to do that, Barabási said it was simple—just bring in the best people at all levels. “This is what has hap­pened,” Aoun said.

While suc­cess may now be accu­rately mea­sured through com­pli­cated math­e­matics thanks to Barabási’s work, what mat­ters most may still be the col­lab­o­ra­tions and links that allow a researcher to sus­tain such work.


]]> 0
Northeastern Breaks Ground on State-​​of-​​the-​​Art Science and Engineering Complex Mon, 24 Feb 2014 14:32:47 +0000 North­eastern Uni­ver­sity and the city of Boston opened a new chapter in their long part­ner­ship on Friday morning when city offi­cials joined uni­ver­sity leaders to break ground on Northeastern’s state-​​of-​​the-​​art Inter­dis­ci­pli­nary Sci­ence and Engi­neering Com­plex on Columbus Avenue.

In his remarks, North­eastern Uni­ver­sity Pres­i­dent Joseph E. Aoun told the hun­dreds of people in atten­dance, including Boston Mayor Marty Walsh, Boston City Coun­cilor Tito Jackson, and state Rep. Jef­frey Sanchez of Jamaica Plain, that the new com­plex would ben­efit the North­eastern com­mu­nity as well as the entire city.

“You can look at this com­plex from dif­ferent per­spec­tives,” Aoun said. “Yes, it is going to serve the stu­dents. Yes, it is going to serve our fac­ulty. And yes, it is going to serve the com­mu­nity. But more impor­tantly it is going to bring every con­stituency together.”

Hundreds of members of the Northeastern University and surrounding communities gathered at the ceremonial groundbreaking for the Interdisciplinary Science and Engineering Complex (ISEC), on Columbus Avenue. Northeastern President Joseph E. Aoun joined by new Boston Mayor Marty Walsh in the ceremony. Photo by Brooks Canaday.

The 220,000-square-foot research and edu­ca­tional space is part of Northeastern’s ongoing effort to expand its capacity to engage in path-​​breaking research across dis­ci­plines. Sched­uled to open in 2016, it will include wet and dry lab facil­i­ties, edu­ca­tional lab­o­ra­to­ries, class­room space, and offices for fac­ulty and grad­uate students.

“A sci­ence com­plex of this scale has the chance to be a shining example of the best Boston has to offer,” Walsh said in his remarks. “This facility will attract some of the world’s best minds in the most cut­ting edge fields of research. The work they will do will change the world in ways we can’t even imagine yet.”

The ISEC will be the first pri­vate research devel­op­ment in Rox­bury and is expected to create more than 600 jobs during the con­struc­tion phase and an addi­tional 700 jobs after the com­plex opens.

“The future of our uni­ver­sity is immi­nently linked to the well being of the Rox­bury com­mu­nity and Boston as a whole,” Aoun said. “You need us. But more impor­tantly, we need you.”

The six-​​story, LEED-​​certified facility will fea­ture cutting-​​edge sci­en­tific equip­ment and lab space, both of which will be shared by researchers from Northeastern’s Col­lege of Sci­ence, Bouvé Col­lege of Health Sci­ences, Col­lege of Engi­neering, and Col­lege of Com­puter and Infor­ma­tion Sci­ence.

“The most sig­nif­i­cant word in the name of this building is inter­dis­ci­pli­nary,” Stephen W. Director, provost and senior vice pres­i­dent for aca­d­emic affairs, said in his remarks. “Our research focuses on finding solu­tions to the global chal­lenges in the areas of health, sus­tain­ability, and secu­rity. Solu­tions to these chal­lenges require the col­lab­o­ra­tion of many minds working together in many fields.”

Miles Graham, a seventh-​​grader at the Match Charter Public School in Boston, said the new facility would help Boston youth realize their dreams of becoming suc­cessful sci­en­tists and engi­neers. “This new building rep­re­sents a bigger and better oppor­tu­nity for Boston’s youth,” he explained. “This is how dreams become real.”

Michael Karolewski, comp­troller for the North­eastern Stu­dent Gov­ern­ment Asso­ci­a­tion, noted that the new com­plex would offer stu­dents more research oppor­tu­ni­ties than ever before.

“It’s hard to believe, but there will be even more oppor­tu­ni­ties for stu­dents like me to learn in their own dis­ci­plines and have the poten­tial to com­mu­ni­cate across these other dis­ci­plines,” said Karolewski, DMSB’16.

Walsh was par­tic­u­larly excited about the project’s plan to con­struct a unique pedes­trian bridge over the MBTA Orange Line, com­muter rail, and Amtrak tracks. The bridge—similar to New York City’s “Highline”—will con­nect two dis­tinct sec­tions of Northeastern’s campus and bol­ster the university’s strong ties to the Rox­bury and Fenway neighborhoods.

“Building bridges is what uni­ver­si­ties should be all about,” Walsh said.

Photo courtesy of Payette.

The site’s devel­op­ment pro­vides an oppor­tu­nity to strengthen the Columbus Avenue cor­ridor, improve pedes­trian con­nec­tions, and create new open space and streetscape ameni­ties to be shared with the sur­rounding com­mu­nity. The project rep­re­sents an invest­ment by the uni­ver­sity of about $225 million.

The new sci­ence com­plex is a key part of Northeastern’s Insti­tu­tional Master Plan to strengthen ties with the local com­mu­nity and the city. Uni­ver­sity offi­cials devel­oped the plan over the past two years in col­lab­o­ra­tion with fac­ulty, stu­dents, staff, city plan­ners, and campus neigh­bors. The Boston Rede­vel­op­ment Authority approved the plan on Nov. 14, 2013.

North­eastern has increased its annual research funding by more than 100 per­cent since 2006, and it has received more than $98 mil­lion in external research funding in 2013. The uni­ver­sity is also diver­si­fying its research funding by delib­er­ately increasing sup­port from phil­an­thropic and cor­po­rate sources, not just gov­ern­ment grants.


]]> 0