Pendharkar P.C.,Information Systems
Engineering Applications of Artificial Intelligence | Year: 2010
In this paper, we propose a software defect prediction model learning problem (SDPMLP) where a classification model selects appropriate relevant inputs, from a set of all available inputs, and learns the classification function. We show that the SDPMLP is a combinatorial optimization problem with factorial complexity, and propose two hybrid exhaustive search and probabilistic neural network (PNN), and simulated annealing (SA) and PNN procedures to solve it. For small size SDPMLP, exhaustive search PNN works well and provides an (all) optimal solution(s). However, for large size SDPMLP, the use of exhaustive search PNN approach is not pragmatic and only the SA-PNN allows us to solve the SDPMLP in a practical time limit. We compare the performance of our hybrid approaches with traditional classification algorithms and find that our hybrid approaches perform better than traditional classification algorithms. © 2009 Elsevier Ltd. All rights reserved. Source
Pendharkar P.C.,Information Systems
Expert Systems with Applications | Year: 2012
We consider game-theoretic principles for design of cooperative and competitive (non-cooperative self-interested) multi-agent systems. Using economic concepts of tâtonnement and economic core, we show that cooperative multi-agent systems should be designed in games with dominant strategies that may lead to social dilemmas. Non-cooperative multi-agent systems, on the other hand, should be designed for games with no clear dominant strategies and high degree of problem complexity. Further, for non-cooperative multi-agent systems, the number of learning agents should be carefully managed so that solutions in the economic core can be obtained. We provide experimental results for the design of cooperative and non-cooperative MAS from telecommunication and manufacturing industries. © 2011 Elsevier Ltd. All rights reserved. Source
With over 130,000 square feet of hands-on makerspaces, MIT has more of these facilities on its campus than anywhere in the world. Yet, according to findings from a student-wide survey conducted last summer, the top two places where MIT students make things are in their dorm rooms and off-campus. The reason? Students face too many barriers when trying to use MIT’s expansive maker infrastructure. Students across the Institute need access to makerspaces not only for their studies but to work on personal and entrepreneurial projects. Rebecca Li, a junior majoring in mechanical engineering, notes that students often have to “hijack a club or lab’s machine shop access, pay many different membership fees, or stumble into little known shops like MITERS [MIT Electronics Research Society] and Maker Works.” According to Li, who helps manage the MITERS space as their facilities manager, the only way to find out about these options is effectively through referral. “If you are not involved in some builder club, lab, or group, you are unlikely to hear about all of the shops or what you have the ability to access,” she says. In Aaron Ramirez’s experience as a PhD student, new arrivals to MIT have little idea of what resources exist on campus and are often lost as to where to begin and where to turn to for help. “Many of the incoming students, including myself long ago, never had exposure to or training on machine tools, rapid prototyping equipment, or instrumentation. These tools are essential for engineering and for obtaining and improving your skills as an engineer — it’s not just a hobby for us, it’s our life and passion, and necessary for our development,” he says. “Students have to put in a decent amount of effort to finally get access to the workshops, which is not trivial when you consider how much other stuff they have to do already, especially as a freshmen!” No one understands these hindrances more than Marty Culpepper, MIT alumnus, professor of mechanical engineering, and the recently appointed MIT “Maker Czar.” He acknowledges that it can take a student up to nine months to get into a space and build anything, a startling statistic that doesn’t sit well with him. “There are reasons why students have difficulty getting into these spaces,” Culpepper says. “One barrier is knowing where everything is. Another is, how do you get trained, whom do you contact to get trained, and once you’ve been trained, do other shops know you’ve got that skill set? How do you pay for things if you need to pay for them? All of these things stacked on top of each other make it very difficult not only for students to get things done, but for faculty and their research to get done.” In an effort to address the situation, Culpepper has been charged with determining the best ways to increase student access to campus makerspaces, which in part means breaking down these barriers in a myriad of ways. “We know that MIT students want to design and build things. We know this is important to their education and to their desire to start companies. We are here to fix the problem and the first step to doing that is with Mobius,” says Culpepper. Designed to help the MIT community navigate a complex making system, the Mobius app (shown here), released March 4, enables users to search through the vast array of makerspaces and equipment on campus. (Christine Daniloff/MIT) Designed to help the MIT community navigate a complex making system, the Mobius app, released March 4, enables users to search through the vast array of makerspaces and equipment on campus. Developed in partnership with students, shop managers, alumni, and MIT’s Information Systems and Technology office, the app is the first of its kind. It was realized through the support of the Lord Foundation of Massachusetts and MIT alumni Colin and Erika Angle, who recognized the potential of Mobius to transform the maker experience for MIT students. Available to download for iOS (with an Android version in development), Mobius will match users’ needs to maker resource availability as well as assist technical staff in managing their shops and improving student communications and interactions. Students who use the app will be able to locate that laser cutter or mill they need for their project from the convenience of their mobile device, saving them time spent searching for information that is not available online. Using the app also eliminates the step of having to call or walk to each space for shop hours, policies, and training protocols. Li, who along with Ramirez worked with Culpepper to design the app, says that Mobius “will let people uncover which makerspace is both open and has the right machines for them.” She continues, “Knowing where certain machines are on campus, or even that they exist, will help people get things done and make more things. The most important aspect of the app is the elimination of the unknown of whom to contact or whom to ask for access. People will feel more confident knowing that they are talking to the right person who can get them help, instead of playing email ping pong as their project gathers dust.” Other key features of Mobius include the ability to pay for materials, machine time, and access fees directly through the app. Additionally, shop managers will be able to check a user’s abilities with a built-in endorsement and flagging system, potentially allowing someone already skilled at machining and endorsed by another shop to fast-track their training to gain access in a different facility. In the future, users in turn will be able to rate their experience with a particular shop, providing helpful tips and advice to others in the maker community. For the last two years, Culpepper has spent his days learning about all of MIT’s makerspaces and visiting other universities across the country to explore theirs. The experience was eye-opening, leading to the realization that while MIT boasted the most makerspace square footage, its utilization rate was underwhelming. Culpepper is determined to lead the future of making at MIT and beyond, with new technologies such as Mobius and other activities of the recently launched Project Manus. Initiated in October 2015 by MIT Provost Martin Schmidt and housed within the MIT Innovation Initiative, Project Manus will build capacity in MIT’s makerspaces and foster the maker communities that will create the gold standard in next-generation academic maker systems. Lessons learned on the MIT campus — including Mobius — will be shared widely so students and staff at other universities can take advantage. “Today marks a very important first step in a journey to provide our community with seamless access to the vast maker resources on our campus, and to bring greater coordination of all these resources. In Mobius, and the other elements of Project Manus, Professor Culpepper has provided us with an exciting vision for the future and a roadmap to get there. I am grateful to Professor Culpepper and the MIT Innovation Initiative for advancing this critical effort to strengthen our innovation ecosystem,” said Provost Schmidt upon the app’s release. Culpepper, whose work is just beginning, remarks, “Mobius is just one of five major programs that Project Manus is working on to improve the way MIT supports students and makerspace staff. I invite everyone to visit the Project Manus web site to learn more.”
Abstract: Healthcare practitioners may one day be able to physically screen for breast cancer using pressure-sensitive rubber gloves to detect tumors, owing to a transparent, bendable and sensitive pressure sensor newly developed by Japanese and American teams. Conventional pressure sensors are flexible enough to fit to soft surfaces such as human skin, but they cannot measure pressure changes accurately once they are twisted or wrinkled, making them unsuitable for use on complex and moving surfaces. Additionally, it is difficult to reduce them below 100 micrometers thickness because of limitations in current production methods. To address these issues, an international team of researchers led by Dr. Sungwon Lee and Professor Takao Someya of the University of Tokyo's Graduate School of Engineering has developed a nanofiber-type pressure sensor that can measure pressure distribution of rounded surfaces such as an inflated balloon and maintain its sensing accuracy even when bent over a radius of 80 micrometers, equivalent to just twice the width of a human hair. The sensor is roughly 8 micrometers thick and can measure the pressure in 144 locations at once. The device demonstrated in this study consists of organic transistors, electronic switches made from carbon and oxygen based organic materials, and a pressure sensitive nanofiber structure. Carbon nanotubes and graphene were added to an elastic polymer to create nanofibers with a diameter of 300 to 700 nanometers, which were then entangled with each other to form a transparent, thin and light porous structure. "We've also tested the performance of our pressure sensor with an artificial blood vessel and found that it could detect small pressure changes and speed of pressure propagation," says Lee. He continues, "Flexible electronics have great potential for implantable and wearable devices. I realized that many groups are developing flexible sensors that can measure pressure but none of them are suitable for measuring real objects since they are sensitive to distortion. That was my main motivation and I think we have proposed an effective solution to this problem." ### This work was conducted in collaboration with the research group of Professor Zhigang Suo at Harvard University, USA. Collaborating institutions Osaka University Harvard University, USA Funding Japan Science and Technology Agency (JST) Exploratory Research for Advanced Technology (ERATO) Someya Bio-Harmonized Electronics Project About University of Tokyo The University of Tokyo is Japan's leading university and one of the world's top research universities. The vast research output of some 6,000 researchers is published in the world's top journals across the arts and sciences. Our vibrant student body of around 15,000 undergraduate and 15,000 graduate students includes over 2,000 international students. Find out more at www.u-tokyo.ac.jp/en/ or follow us on Twitter at @UTokyo_News_en. For more information, please click Contacts: Research contact Professor Takao Someya Department of Electrical Engineering and Information Systems Graduate School of Engineering The University of Tokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo 113-8656, Japan Tel: +81-3-5841-0411/6756 Fax: +81-3-5841-6709 Press officer contact Graduate School of Engineering Public Relations Office The University of Tokyo The University of Tokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo 113-8656, Japan Tel: 03-5841-1790 Fax: 03-5841-0529 If you have a comment, please us. Issuers of news releases, not 7th Wave, Inc. or Nanotechnology Now, are solely responsible for the accuracy of the content.
Rochester, NY — Cloud computing software has brought many changes to the business landscape and, currently, the implementation of such a service is common. New research sheds light on the growing competition between two widely used software models — Software-as-a-Service (SaaS), and Modified off-the-Shelf (MOTS) software. “The key factor that drives competitive business strategies in this highly aggressive market is the provider’s pricing scheme,” says Abraham Seidmann, Xerox Professor of Computers and Information Systems and Operations Management from the University of Rochester’s Simon Business School. SaaS is available online on-demand, which allows businesses to increase production with fewer people. This scalability reduces investment on technology infrastructure, as well as easily maintaining access to important information with little to no upfront spending. SaaS is used in a number of common business areas with organizations, such as Workday, which offers cloud-based enterprise-level software solutions for human resources and financial management, QuickBooks offering cloud-based accounting solutions and DialogsCMS delivering enterprise content management. The typical in-house MOTS systems provide some API’s (application program interface) with access to the source code of the underlying software so it can be customized and better integrated to the business. Cerner is an example of a health care technology vendor that develops customized Electronic Medical Records (EMR) to create a “custom fit” between the software and the needs of the medical institution. This vendor recognizes the need for customization and enhanced functionality, while a leading competitor, such as EPIC, seems to be far more limited in that respect. On the other hand, most SaaS systems provide limited customization options, because they are operating in a multitenancy environment. Multiple customers share the same application, running on the same operating system, hardware and data-storage mechanism. This is how SaaS attains economies of scale, but as a result, users might incur significant lack-of-fit and integration costs. According to Seidmann, “Paying more while not getting your exact business integration needs is called “lack-of-fit” cost. When lack-of-fit costs decrease — when new industry standards are adopted, for example — SaaS systems need to reduce prices to gain market share. However, when lack-of-fit costs are expected to increase, MOTS software will be more competitive, because it’s easier to modify the source code to meet specific functional needs.” In their paper, “Analyzing Software as a Service (SaaS) with Per-Transaction Charges,” Seidmann and his co-author Dan Ma, Associate Professor Information Systems from Singapore Management University, built a game theory model to explore competitive pricing strategies of SaaS and MOTS platforms on a per-transaction basis to determine where and how each service modality gives end-users more value for their money. The co-authors identified three different qualities to measure the pricing strategies between competitors: Based on their analysis, the researchers offer three top strategic recommendations: Now that the cloud computing software trend seems unstoppable in many markets, major players in the competitive on-demand software game need to adapt to the changing times and offer both SaaS and MOTS options. The trend is already evident for companies, such as SAP, Microsoft and Oracle who feel the pressure by newcomers to the cloud space and, hence, offer both versions — hoping to convert a one-time sale into a perpetual income stream.