1,736,169 research outputs found
Conversational adjustments in interaction (Pham & Viswanathan, 2024)
No full textPurpose: We examined which measures of complexity are most informative when studying language produced in interaction. Specifically, using these measures, we explored whether native and nonnative speakers modified the higher level properties of their production beyond the acoustic–phonetic level based on the language background of their conversation partner.Method: Using a subset of production data from the Wildcat Corpus that used Diapix, an interactive picture matching task, to elicit production, we compared English language production at the dyad and individual level across three different pair types: eight native pairs (English–English), eight mixed pairs (four English–Chinese and four English–Korean), and eight nonnative pairs (four Chinese–Chinese and four Korean–Korean).Results: At both the dyad and individual levels, native speakers produced longer and more clausally dense speech. They also produced fewer silent pauses and fewer linguistic mazes relative to nonnative speakers. Speakers did not modify their production based on the language background of their interlocutor.Conclusions: The current study examines higher level properties of language production in true interaction. Our results suggest that speakers’ productions were determined by their own language background and were independent of that of their interlocutor. Furthermore, these demonstrated promise for capturing syntactic characteristics of language produced in true dialogue.Supplemental Material S1. Transcription guidelines.Supplemental Material S2. Transcription sample (Levels 1–3).Supplemental Material S3. Supplemental analyses (Levels 1–3).Pham, C. T., & Viswanathan, N. (2024). Studying conversational adjustments in interaction: Beyond acoustic phonetic changes, Journal of Speech, Language, and Hearing Research, 67(1), 196–210. https://doi.org/10.1044/2023_JSLHR-23-00268</p
Characterization of TLB and page allocation behavior on modern processors
No full textVirtual memory support is prevalent in most modern processors and is facilitated through Translation Lookaside Buffers (TLBs) which play a major role in the overall system performance. TLB misses are costly since they require multiple high latency memory references to walk the page table and locate the desired Virtual Page Number (VPN) - Physical Page Number (PPN) mapping. This study improves TLB hit rates by taking advantage of any contiguity present in the pages allocated by the Operating System (OS). By contiguity we refer to cases where consecutive VPNs are mapped to consecutive PPNs. Traditionally, OSs use large or superpages to collapse hundreds of such contiguous entries, thereby using one TLB entry to represent them rather than hundreds of entries they would normally require. Unfortunately due to implementation complexities superpaging has not been universally successful in reducing TLB pressure. We show, however, that even without explicit superpaging, various OS virtual memory allocation activities lead to intermediate levels of contiguity that may be exploited to coalesce TLB entries and significantly improve hit rates. We verify the presence of contiguity by running benchmarks on a real system and checking the page allocations of the OS. The OS page allocation schemes depend on memory pressure and memory defragmentation daemons. Further, we find an average contiguity of 30 pages over all the benchmarks and configurations with superpaging turned on and about 10 with superpaging turned off. To verify the performance of a Coalesced TLB we have implemented a fully associative TLB with variable size and Least Recently Used (LRU) replacement policy. Our results show an average hit rate improvement of 25% by adding an 8-16 entry fully associative Coalesced TLB. The Coalesced TLB further needs no complex hardware to implement, hence providing to a low cost means to reduce miss rates.M.S.Includes bibliographical referencesby Viswanathan Vaidyanatha
Sustainable product and market development for subsistence marketplaces: Creating educational initiatives in radically different contexts
No full textDeveloping products and business processes to serve subsistence marketplaces (or the roughly 4 billion poor around the world referred to as the bottom of the pyramid) is a significant challenge for businesses. Despite the importance of subsistence marketplaces, most product development educational curricula have been focused on relatively resource-rich and literate consumers and markets. We teach an innovative year-long product development course which includes an international immersion experience and which covers a broad spectrum of learning from understanding poverty, to consumer behavior, to product development and engineering design specifically for subsistence consumers. This unique course represents a pioneering effort to focus attention and create knowledge about product development, marketing, management, and engineering practices for subsistence marketplaces. Our two-semester course sequence for graduate-level students in a variety of business and engineering disciplines and industrial design combines in-class pedagogy with experiential learning and results in useful and marketable product concepts and prototypes. Working on projects with multinational companies or startups, students identify an opportunity of general need, conduct field market research to better understand subsistence consumer needs and contexts through an international immersion experience, develop a product concept, convert the concept to a workable prototype, and develop a manufacturing plan, marketing strategy, and overall business plan for the product. Overlaying the content found in a typical new product development lab course we develop a contextual understanding of subsistence marketplaces, setting the stage for new product development. A central aspect of the learning experience is travel to subsistence markets for actual immersion in the context and to conduct market research. Our course is at the confluence of two of the most important issues facing humanity, subsistence and sustainability. Lessons learned here can also be extended to other radically different contexts, such as future scenarios involving severe energy shortages or climate change consequences. Such educational initiatives provide challenging learning experiences in preparing students for the unique demands of the 21st century. © 2011 Product Development and Management Association.APPLE LE, 1988, J PROD INNOVAT MANAG, V5, P70, DOI 10.1111-1540-5885.510070; ASHBY M, 2003, MATER TODAY, V6, P24, DOI 10.1016-S1369-7021(03)01223-9; Cardozo RN, 2002, J PROD INNOVAT MANAG, V19, P4, DOI 10.1016-S0737-6782(01)00116-3; CHICK A, 1997, J SUSTAINABLE PRODUC, V1, P53; Donaldson KM, 2006, RES ENG DES, V17, P135, DOI 10.1007-s00163-006-0017-3; Ehrenreich B., 2002, NICKEL DIMED; Eppinger S.D., 2002, DESIGN MANAGEMENT J, V13, P58; GESCHKA H, 1986, J PROD INNOVAT MANAG, V3, P48, DOI 10.1016-0737-6782(86)90043-3; GREEN M, 2006, P ASME DES ENG TECHN; HAMMOND A. L., 2007, INNOVATIONS, V2, P147, DOI [10.1162-itgg.2007.2.1-2.147, DOI 10.1162-ITGG.2007.2.1-2.147]; HANNUKAINEN P, 2006, P ASME DES ENG TECHN; Hargadon A, 2000, HARVARD BUS REV, V78, P157; HAUSER JR, 1988, HARVARD BUS REV, V66, P63; HERSTATT C, 1992, J PROD INNOVAT MANAG, V9, P213, DOI 10.1016-0737-6782(92)90031-7; HORAN J, 2004, ONE PAGE BUSINESS PL; Murcott S., 2007, J INT DEV, V19, P123, DOI 10.1002-jid.1353; Prahalad CK, 2002, HARVARD BUS REV, V80, P48; PRAHALAD CK, 2005, FORT BOTT PYR ER POV; PUGH P, 1991, TOTAL DESIGN INTEGRA; Rodriguez J, 2006, INTERACT COMPUT, V18, P956, DOI 10.1016-j.intcom.2006.05.007; Sahlman WA, 1997, HARVARD BUS REV, V75, P98; Schumacher Ernest F., 1973, SMALL IS BEAUTIFUL E; *SME, 2003, PLAST INJ MOLD; *SME, 2004, FUND MAN PROC SAMPL; Sridharan S, 2008, J CONSUM MARK, V25, P455, DOI 10.1108-07363760810915671; Stone R. B., 2000, Design Studies, V21, DOI 10.1016-S0142-694X(99)00003-4; TAGUCHI J, 1990, HARVARD BUS REV, V68, P65; Tybout JR, 2000, J ECON LIT, V38, P11, DOI 10.1257-jel.38.1.11; Ulrich K., 2007, PRODUCT DESIGN DEV; Viswanathan M, 2005, J MARKETING, V69, P15, DOI 10.1509-jmkg.69.1.15.55507; Viswanathan M., 2009, IVEY BUSINESS J MAR; Viswanathan M, 2009, J MACROMARKETING, V29, P406, DOI 10.1177-0276146709345620; Viswanathan M, 2008, EDUC ASIA PAC REG-IS, V12, P1, DOI 10.1007-978-1-4020-5769-4; Viswanathan M., 2007, PRODUCT MARKET DEV S, P1; Viswanathan M., 2007, PRODUCT MARKET DEV S, P212
Genetic algorithms for uncapacitated network design
Full text linkThesis (M.S.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 1995.Includes bibliographical references (leaves 124).by Viswanathan Lakshmi.M.S
- …
