"uuid","repository link","title","author","contributor","publication year","abstract","subject topic","language","publication type","publisher","isbn","issn","patent","patent status","bibliographic note","access restriction","embargo date","faculty","department","research group","programme","project","coordinates"
"uuid:a41607f7-6a0c-4a0a-8895-5aa963c86d6f","http://resolver.tudelft.nl/uuid:a41607f7-6a0c-4a0a-8895-5aa963c86d6f","Implementing the Reordered PageRank Algorithm in Giraph","Oostenbrink, J.","Van Gijzen, M.B. (mentor); De Weerdt, M.M. (mentor)","2015","PageRank, a method to rank web pages objectively and mechanically, models a random web surfer. The PageRank problem is most easily solved iteratively, using the power method. In this paper the reordered PageRank algorithms are discussed. These algorithms (proposed by A. N. Langville and C. D. Meyer in ""A reordering for the PageRank problem"") see the PageRank problem as a linear system of equations and begin by reordering the input Graph/matrix. This way only a smaller problem has to be solved. A disadvantage is that it does take a few extra steps to gain the PageRank values from the solution to this smaller problem. We've developed a suitable stopping condition for these algorithms. However, numerical experiments indicate that this stopping condition is much stricter than the stopping condition for the power method. The reordered PageRank algorithms and the power method have been implemented in Giraph, an open source version of Pregel. Pregel and Giraph are frameworks for solving large graph problems distributively in a vertex centred manner. Because of some of the bugs and features in Giraph (and the inherent complexity of the reordered PageRank algorithms), implementing the reordered PageRank algorithms is much more complicated than implementing the power method. The reordered PageRank algorithms are not faster than the power method in Giraph. Even when accounting for the difference in stopping condition the power method is much faster than the reordered PageRank algorithms.","PageRank; Reordering; Power Method; Power Iteration; Giraph","en","bachelor thesis","","","","","","","","","Electrical Engineering, Mathematics and Computer Science","Applied mathematics","","","",""
"uuid:4d1d2323-8c1f-4a98-a6ba-f44b3e25ba9e","http://resolver.tudelft.nl/uuid:4d1d2323-8c1f-4a98-a6ba-f44b3e25ba9e","Google PageRank and Reduced-Order Modelling","De Looij, H.R.","Budko, N.V. (mentor)","2013","Google PageRank is designed to determine the importance of a webpage. To do so, one needs to compute an eigenvector of the Google matrix. We show that this vector can also be found by solving a linear system. Additionally, an adjustment to the PageRank model will be made by considering one of the parameters to be a stochastic variable. We will develop a reduced-order algorithm which is able to approximate the expected PageRank vector.","Google; PageRank; reduced-order model","en","bachelor thesis","","","","","","","","2013-08-06","Electrical Engineering, Mathematics and Computer Science","Delft Institute of Applied Mathematics","","Applied Mathematics","",""
"uuid:16c5c2ba-12eb-429e-9a95-9ed2eeea668e","http://resolver.tudelft.nl/uuid:16c5c2ba-12eb-429e-9a95-9ed2eeea668e","The second eigenvector of the Google matrix and its relation to link spamming","Sangers, A.","Van Gijzen, M.B. (mentor); Vuik, C. (mentor); Spandaw, J.G. (mentor); Dubbeldam, J.L.A. (mentor)","2012","Google PageRank attempts to return the best ranking of websites when searching on the web. To find this ranking, Google introduces a Markov matrix to model the behaviour of internet users. We find that using irreducible closed subsets is an effective way to unfairly increase the PageRank of a website (perform link spamming) and this can be detected when looking at the second eigenvector of the Google matrix.","Google; PageRank; link spamming; eigenvector","en","bachelor thesis","","","","","","","","2012-10-31","Electrical Engineering, Mathematics and Computer Science","Applied mathematics","","","",""
"uuid:295511ed-dd50-496e-bcc3-33c4b02ba0dc","http://resolver.tudelft.nl/uuid:295511ed-dd50-496e-bcc3-33c4b02ba0dc","Google's PageRank: De tweede eigenwaarde en een variable sprongkans","Baksteen, T.","Van Gijzen, M.B. (mentor)","2012","Google maakt gebruik van PageRank om webpagina's te rangschikken naar hoe belangrijk ze zijn. Het berekenen van de PageRank vereist lineaire algebra. Ook komt Markov theorie hier aan bod. Één van de parameters is een zekere alfa. Deze alfa wordt door velen constant gekozen. Ik heb in dit rapport onderzoek gedaan naar de gevolgen van het variabel maken van deze alfa.","Google; PageRank; Tweede eigenwaarde; Web; Pagina","nl","bachelor thesis","","","","","","","","2012-04-27","Electrical Engineering, Mathematics and Computer Science","Numerieke wiskunde","","","",""
"uuid:935f65ff-42bc-43dd-9e79-ec5fee6f48f3","http://resolver.tudelft.nl/uuid:935f65ff-42bc-43dd-9e79-ec5fee6f48f3","Efficiente benadering van Google's PageRank","Berkhof, E.","Vuik, C. (mentor)","2010","Google maakt gebruik van de PageRankmethode om een lijst van pagina's te maken van belangrijk naar onbelangrijk. Om deze lijst te krijgen uit een enorme grote matrix gebruiken ze de eigenvector die hoort bij de eigenwaarde 1, die deze matrix altijd heeft. Om deze eigenvector te bepalen wordt de Powermethode gebruikt. In dit verslag probeer ik een efficientere methode te vinden om deze eigenvector te bepalen. Voor kleinere matrices kunnen andere methoden efficienter zijn, maar voor zo'n enorme matrix werkt de Powermethode het efficientst.","google; PageRank","nl","bachelor thesis","","","","","","","","2010-09-06","Electrical Engineering, Mathematics and Computer Science","Applied mathematics","","Technische Wiskunde","",""
"uuid:0483fd00-117d-43a3-b61f-6ce8a178e709","http://resolver.tudelft.nl/uuid:0483fd00-117d-43a3-b61f-6ce8a178e709","The PageRank Problem","Den Besten, M.","Van Gijzen, M.B. (mentor)","2010","The thesis is the result of a bachelor research project about Google's PageRank. An analysis of the hyperlink structure of the World Wide Web is made and a model for web surfing studied. Based on this model, some standard methods to compute the PageRank of web pages is investigated. Special attention is given to computing PageRanks by using linear systems. In this respect the IDR(s) method is applied to achieve an efficient computation. Several numerical experiments are performed in order to compare the efficiency of different methods in computing PageRanks.","Google; PageRank; IDR(s); Power method; Jacobi iteration; Gauss-Seidel iteration; Markov chain","en","bachelor thesis","","","","","","","","2016-04-01","Electrical Engineering, Mathematics and Computer Science","Applied mathematics","","","",""