1 / 11

PageRank

PageRank. Un Motor de Búsqueda. “ obama ”. PageRank Model: Final Version. The Web: a directed graph. Edges ( links ). Vertices ( pages ). f. a. e. b. d. c. Input Structure. 41.5 million edges 5.4 million nodes document-with-link document-linked. Step 1. Dictionary Encode Links.

zenda
Télécharger la présentation

PageRank

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. PageRank

  2. Un Motor de Búsqueda

  3. “obama”

  4. PageRank Model: Final Version • The Web: a directed graph Edges (links) Vertices (pages) f a e b d c

  5. Input Structure • 41.5 million edges • 5.4 million nodes document-with-link document-linked

  6. Step 1. Dictionary Encode Links • Strings difficult to fit in memory • Encode strings as OIDs (object ids = integers) • Input line: http://es.dbpedia.org/resource/Ciencia_ficción http://es.dbpedia.org/resource/Robot • Output line: • 52673 • Dictionary: • http://es.dbpedia.org/resource/Ciencia_ficción … 52673 http://es.dbpedia.org/resource/Robot … • OIDCompress -i[folder]/page_links_es.tsv.gz -igz -o [folder]/page_links_es.oid.gz -ogz -d [folder]/page_links_es.dict.gz -dgz

  7. Step 2. Write PageRank Algorithm • PageRankGraph.rankGraph(int[][] graph) • int[] out = graph[i]; • out contains the nodes linked from node i • it might be empty or null if node i doesn’t link to anything! • two rank vectors: rank[graph.length], nextRank[graph.length] • initial rank values set as 1d / graph.length • run ITERS number of iterations • compute edge-invariant rank once per iteration (red and blue) • need to keep track of sum of ranks of nodes with no outlinksfrom prev. round • for each node (orange) • split it’s rank[] by the number of outlinks it has, and add the result to the nextRank[] of each node it links to • the sum of the ranks after each round should be very very close to 1 • test on –idata/test-graph.txt –o data/test-data.txt

  8. Step 3. Rank full data • Run ranking -i [folder]/page_links_es.oid.gz -igz -o [folder]/page_ranks_es.oid.gz –ogz • Sort by rank -i [folder]/page_ranks_es.oid.gz -igz -o [folder]/page_ranks_es_s.oid.gz –ogz • Decompress the file -d [folder]/page_links_es.dict.gz -dgz -i [folder]/page_ranks_es_s.oid.gz -igz -n 0 -o [folder]/page_ranks_es_s.tsv.gz -ogz

  9. Course Marking • 45% for Weekly Labs (~3% a lab!) • 35% for Final Exam • 20% for Small Class Project

  10. Class Project • Done in pairs (Except Alejandro :P) • Goal: Use what you’ve learned to do something cool (basically) • Expected difficulty: More than a lab’s worth • But from scratch / without my help! • Marked on: Difficulty, appropriateness, scale, good use of techniques, presentation, coolness • Ambition is appreciated, even if you don’t succeed: feel free to bite off more than you can chew! • Process: • Pair up (default random) by Wednesday • Decide on a topic (by June 9th) or let me assign one  • If you need data or get stuck, I will (try to) help out • Deliverables: 10 minute presentation (June 23rd) & 4-page report

More Related