A while back (2002?) I did a literal translation of the Perl soundex module to VB.Net, to help me learn. Since VB.Net lacks a transliteration operator, I used nested loops over arrays of characters and their replacements, what a mess! It ended up being rediculously slow. Since soundex "knows" the sets of characters at compile time, your method makes a lot more sense, and works much more efficiently.
J.Ja
Discussion on:
View:
Show:
Wikipedia has more information about Soundex: http://en.wikipedia.org/wiki/Soundex including samples in other programming languages. An important thing to note is that the Soundex rules works well for ENGLISH words and may give erroneous results for other non-phonetic languages. More info at the Wiki...
Be aware that Soundex totally falls apart if the initial letters are different. Site and cite for example will not match. Metaphone is much better because it allows for more of the vagarities of English.
Hello!
Do you consider cooperation with dedicated Php developer or dedicated development team based on beneficial terms?
Let me introduce myself, my name is Maxim Slobodyanyuk.
I am a founder and CEO of http://www.ilogos-ua.com (iLogos - outsourcing software development team, based on Ukraine)
We have pretty good PHP developer, his name is Alex. He has 6 years of web-development experience includes: project architecture creation, design of user interfaces, multilayer web-applications. 3 year experience in object-oriented development: creation of database management system, graphics processing applications, web-site processing.
If you are interested in dedicated PHP developer ? feel free to contact me (skype: getstar777, icq: 217580424, gtalk: slobod777).
Do you consider cooperation with dedicated Php developer or dedicated development team based on beneficial terms?
Let me introduce myself, my name is Maxim Slobodyanyuk.
I am a founder and CEO of http://www.ilogos-ua.com (iLogos - outsourcing software development team, based on Ukraine)
We have pretty good PHP developer, his name is Alex. He has 6 years of web-development experience includes: project architecture creation, design of user interfaces, multilayer web-applications. 3 year experience in object-oriented development: creation of database management system, graphics processing applications, web-site processing.
If you are interested in dedicated PHP developer ? feel free to contact me (skype: getstar777, icq: 217580424, gtalk: slobod777).
seems like you have an error in your logic for the second part, There is the same issue being covered twice.
1.Run both words through Soundex
2. If the Soundex encodings are the same the rank will be 4. If the Soundex encodings are not the same, continue to step 3.
3. If the last three characters of the first encoding are found in the second encoding the rank will be 3. Skip to step 7.
4. If the last two characters of the first encoding are found in the second encoding the rank will be 2. Skip to step 7.
5. If the middle two characters of the first encoding are found in the second encoding, the rank will be 2. Skip to step 7.
6. If the second, third, or fourth characters of the first encoding are found within the second encoding add 1 to the rank for each character that matches.
7. If the first character of the first encoding matches the first character of the second encoding, add 1 to the current rank.
if it passes 2 then 3 need not jump to 7, as that would then mean it would equal the result for 2, for my input I would be likely to drop step 2 and just jump into step 3
1.Run both words through Soundex
2. If the Soundex encodings are the same the rank will be 4. If the Soundex encodings are not the same, continue to step 3.
3. If the last three characters of the first encoding are found in the second encoding the rank will be 3. Skip to step 7.
4. If the last two characters of the first encoding are found in the second encoding the rank will be 2. Skip to step 7.
5. If the middle two characters of the first encoding are found in the second encoding, the rank will be 2. Skip to step 7.
6. If the second, third, or fourth characters of the first encoding are found within the second encoding add 1 to the rank for each character that matches.
7. If the first character of the first encoding matches the first character of the second encoding, add 1 to the current rank.
if it passes 2 then 3 need not jump to 7, as that would then mean it would equal the result for 2, for my input I would be likely to drop step 2 and just jump into step 3
- Keyboard Shortcuts:
- Prev
- Next
- Toggle
