Content deleted Content added
Chrism2671 (talk | contribs) |
No edit summary |
||
(2 intermediate revisions by 2 users not shown) | |||
Line 1:
{{WikiProject
{{WikiProject
{{WikiProject Computing|importance=}}
}}
---
Line 132 ⟶ 134:
:Strongly agree. I work with matrix maths a lot, but it's obvious to me that the explanation given in this page is quite hard to understand. It seems to be oriented towards computational efficiency rather than clear exposition. The bit with the emission probabilities placed into a diagonal matrix in particular, really rams the point home that the layout is not so we can understand it! --[[User:Mcld|mcld]] ([[User talk:Mcld|talk]]) 11:05, 2 May 2012 (UTC)
I agree. There's an additional problem; the evidence states 00100 are a [[palindrome]], which is extremely confusing for an algorithm that is supposed to have separate forward & backward steps. Furthermore, the transition matrix is symmetric, which also makes the backwards step more unclear. Should you transpose it? [[User:Chrism2671|Chrism2671]] ([[User talk:Chrism2671|talk]]) 12:45, 6 April 2022 (UTC)
== Merge with BCJR algorithm ==
Line 167 ⟶ 169:
Mouse7mouse9 20:20, 28 January 2014 (UTC) <small><span class="autosigned">— Preceding [[Wikipedia:Signatures|unsigned]] comment added by [[User:Mouse7mouse9|Mouse7mouse9]] ([[User talk:Mouse7mouse9|talk]] • [[Special:Contributions/Mouse7mouse9|contribs]]) </span></small><!-- Template:Unsigned --> <!--Autosigned by SineBot-->
== Forward Probabilities ==
There is already a page on Forward Algorithm. It would be better to reference that instead of repeating. The section on backward probabilities starts:
A similar procedure can be constructed to find backward probabilities.
referring the section on forward probabilities, forcing the reader who's already read the article on the Forward Algorithm to go review above section to ensure they understand the reference ...
|