Content deleted Content added
Tag: |
→Why does this work?: new section |
||
Line 258:
::::::You're a distinguished Professor of Computer Science and an ACM fellow! (GULP) You're more qualified to write this article than me, and I wasn't very familiar with it when I started. Throw in something, and I'll do, too. Starting with the definition in the first sentence. Who says that it can find multiple patterns? The seminal paper listed below just says it finds the first occurrence of a pattern in text... i.e. not even multiple occurrences of one pattern.[[User:Sbalfour|Sbalfour]] ([[User talk:Sbalfour|talk]]) 18:24, 28 September 2019 (UTC)
:::::::Just keep running the algorithm and it will find multiple instances of the same pattern. Use a separate hash table to store the hashes of the patterns and it will be able to find multiple different patterns. The same idea, easily generalized. We need published sources for that of course but they don't have to be in the original paper. The subtlety is that the expected time becomes O(n + sum of lengths of matches), which could be larger than n for some inputs. (Also technically I'm not a distinguished professor; that's a specific job title that I don't currently have.) —[[User:David Eppstein|David Eppstein]] ([[User talk:David Eppstein|talk]]) 18:32, 28 September 2019 (UTC)
== Why does this work? ==
The hash function suggested in [[Rabin–Karp algorithm#Hash function used]] goes over each character. So I don't see how using the hash is faster than just comparing each character [[User:AltoStev|<span style='color: orange;'>'''AltoStev'''</span>]] ([[User Talk:AltoStev|<span style='color: #448cff'>talk</span>]]) 23:00, 28 September 2024 (UTC)
| |||