Detecting phone numbers using sapi - java

This is regarding to the question in Detecting phone numbers using sapi. I used that grammar in the answer.
But it gets numbers with space how can I create grammar get numbers without space?

Short answer - it would be quite difficult to do that, as SAPI is designed to recognize words, which have spaces in between them.
However, it's pretty straightforward to annotate the grammar so that you can find the start & end position of the phone number within the reco result, at which point you can remove the spaces using string replacement.
Alternatively, it's also pretty straightforward to place the phone number in a tag, at which point you can extract the number from the rules structure. (This might be trickier from Java.)

Related

Java String.split() spiralling out of control

I am trying to parse strings (some can be very long, paragraphs) based on white space (spaces, return keys, tabs). Currently using String.split("\\s++"). In the previous project we are updating, we had simply used StringTokenizer. Using String.split("\\s++") works just fine in all our testing and with all our beta testers.
The minute we release it to expanded users, it runs for a while until it soaks up all server resources. From what I've researched, it appears to be catastrophic backtracking. We get errors like:
....was in progress with java.base#11.0.5/java.util.regex.Pattern$GroupHead.match(Pattern.java:4804)
java.base#11.0.5/java.util.regex.Pattern$Start.match(Pattern.java:3619)
java.base#11.0.5/java.util.regex.Matcher.search(Matcher.java:1729)
java.base#11.0.5/java.util.regex.Matcher.find(Matcher.java:746)
java.base#11.0.5/java.util.regex.Pattern.split(Pattern.java:1264)
java.base#11.0.5/java.lang.String.split(String.java:2317)
Users can type some crazy text. What is the best option to parse strings that could be anywhere from 10 characters to 1000 characters long? I am at a brick wall. Been trying different patterns (regex is not my strongest area) for the past 4 days without long term success.
The simple solution if you dont trust the regex is to use a non regex based solution such as ApacheCommons StringUtils#split. Alternatively, its pretty easy to write one yourself.
Keep in mind the difference between using StringTokenizer versus a split function is the tokenizer is lazy. If you were only retrieving a subset of the split results you may be eating up more memory with a split. I would only expect this to be a problem with large strings though.

Regex expression for comma and dash seperated text of items

I do have a Java Web Application, where I get some inputs from the user. Once I got this input I have to parse it and the parsing part depends on what kind of input I'll get. I decided to use the Pattern class of java for some of predefined user inputs.
So I need the last 2 regex patterns:
a)Enumaration:
input can be - A03,B24.1,A25.7
The simple way would be to check if there are a comma in there ([^,]+) but it will end up with a lot of updates in to parsing function, which I would like to avoid. So, in addition to comma it should check if it starts with
letter
minimum 3 letters (combined with numbers)
can have one dot in the word
minimum 1 comma (updated it)
b) Mixed
input can be A03,B24.1-B35.5,A25.7
So all of what Enumuration part got, but with addition that it can have a dash minimum one.
I've tried to use multiple online regex generators but didnt get it correct. Would be much appreciated if you can help.
Here is what I got if its B24.1-B35.5 if its just a simple range.
"='.{1}\\d{0,2}-.{1}\\d{0,2}'|='.{1}\\d{1,2}.\\d{1,2}-.{1}\\d{1,2}.\\d{1,2}'";
Edit1: Valid and Invalid inputs
for a)Enumaration
A03,B24.1,A25.7 Valid
A03,B24.1 Valid
A03,B24.1-B25.1 -Invalid because in this case (enumaration) it should not contain dash
A03 invalid because no comma
A03,B24.1 - Valid
A03 Invalid
for b)Mixed
everything that a enumeration has with addition that it can have dash too.
You can use this regex for (a) Enumeration part as per your rules:
[A-Za-z][A-Za-z0-9]{2,}(?:\.[A-Za-z0-9]{1,})?(?:,[A-Za-z][A-Za-z0-9]{2,}(?:\.[A-Za-z0-9]{1,})?)+
Rules:
Verifies that each segment starts with a letter
Minimum of three letters or numbers [A-Za-z][A-Za-z0-9]{2,}
Optionally followed by decimal . and one or more alphabets and numbers i.e (?:\.[A-Za-z0-9]{1,})?
Same thing repeated, and seperated by a comma ,. Also must have atleast one comma so using + i.e (?:,[A-Za-z][A-Za-z0-9]{2,}(?:\.[A-Za-z0-9]{1,})?)+
?: to indicate non-capturing group
Using [A-Za-z0-9] instead of \w to avoid underscores
Regex101 Demo
For (b) Mixed, you haven't shared too many valid and invalid cases, but based on my current understanding here's what I have:
[A-Za-z][A-Za-z0-9]{2,}(?:\.[A-Za-z0-9]{1,})?(?:[,-][A-Za-z][A-Za-z0-9]{2,}(?:\.[A-Za-z0-9]{1,})?)+
Note that , from previous regex has been replaced with [,-] to allow - as well!
Regex101 Demo
// Will match
A03,B24.1-B35.5,A25.7
A03,B24.1,A25.7
A03,B24.1-B25.1
Hope this helps!
EDIT: Making sure each group starts with a letter (and not a number)
Thanks to #diginoise and #anubhava for pointing out! Changed [A-Za-z0-9]{3,} to [A-Za-z][A-Za-z0-9]{2,}
As I said in the comments, I would chop the input by commas and verify each segment separately. Your domain ICD 10 CM codes is very well defined and also I would be very wary of any input which could be non valid, yet pass the validation.
Here is my solution:
regex
([A-TV-Z][0-9][A-Z0-9](\.?[A-Z0-9]{0,4})?)
... however I would avoid that.
Since your domain is (moste likely) medical software, people's lives (or at least well being) is at stake. Not to mention astronomical damages and the lawyers ever-chasing ambulances. Therefore avoid the easy solution, and implement the bomb proof one.
You could use the regex to establish that given code is definitely not valid. However if a code passes your regex it does not mean that it is valid.
bomb proof method
See this example: O09.7, O09.70, O09.71, O09.72, O09.73 are valid entries, but O09.1 is not valid.
Therefore just get all possible codes. According to this gist there are 42784 different codes. Just load them to memory and any code which is not in the set, is not valid. You could compress said list and be clever about the encoding in memory, to occupy less space, but verbatim all codes are under 300kB on disk, so few MBs max in memory, therefore not a massive cost to pay for a price of people not having left instead of right kidney removed.

convert similar sound word parts

I'm having trouble searching for the right terms here to solve the below problem; I'm sure it's a done thing, I just can't find the right terms to express the problem!
I'm basically trying to create a classifier that will take word comparison outputs (e.g. some outputs from Levenstein distances) and decide whether the words are sufficiently different. An important input would probably be something like a soundex comparison. The trouble I'm having is creating the training set for the algorithm (an SVM in this case). I have a long list of names and I need to mutate them a bit (based on similar sounds within the word).
E.g. John and Jon would be a mutation to make, and I could label this in the test set as being equivalent. John and Johann have sufficiently different sound and letter distance to be considered different.
So I'm kinda asking for is a way to achieve a phoneme variation generator, but need to be able to retain the English lettering structure.
Even simple translation might suffice, like "f" could (sometimes) be replaced by "ph". I'm doing this in Java so any tips in that direction would be great too! Thanks.
EDIT
This is the closest I've come across so far: http://www.isi.edu/natural-language/people/hovy/papers/07IJCAI-spelling-variants.pdf
I'm just thinking aloud.
Rule-based: Apply a rule-based system where you could use standard substitution rules such as 'ph' for 'f', and insertion rules such as insert an h between a vowel and a consonant.
Character n-gram alignment:
Use a word alignment tool such as Giza++ to align character n-grams from parallel corpora such as Europarl. I guess you would be able to find interesting word spelling variations such as "house", "haus" etc. You can play with various values of n.
Bootstraping character n-gram alignment with rule-based: You might also want to use a combination of the two, in which you could, in principle, boost the probabilities of some alignments by using a set of external rules and heuristics.

Java & Regex: Matching a substring that is not preceded by specific characters

This is one of those questions that has been asked and answered hundreds of times over, but I'm having a hard time adapting other solutions to my needs.
In my Java-application I have a method for censoring bad words in chat messages. It works for most of my words, but there is one particular (and popular) curse word that I can't seem to get rid of. The word is "faen" (which is simply a modern slang for "satan", in the language in question).
Using the pattern "fa+e+n" for matching multiple A's and E's actually works; however, in this language, the word for "that couch" or "that sofa" is "sofaen". I've tried a lot of different approaches, using variations of [^so] and (?!=so), but so far I haven't been able to find a way to match one and not the other.
The real goal here, is to be able to match the bad words, regardless of the number of vowels, and regardless of any non-letters in between the components of the word.
Here's a few examples of what I'm trying to do:
"String containing faen" Should match
"String containing sofaen" Should not match
"Non-letter-censored string with f-a#a-e.n" Should match
"Non-letter-censored string with sof-a#a-e.n" Should not match
Any tips to set me off in the right direction on this?
You want something like \bf[^\s]+a[^\s]+e[^\s]+n[^\s]\b. Note that this is the regular expression; if you want the Java then you need to use \\b[^\\s]+f[^\\s]+a[^\\s]+e[^\\s]+n[^\\s]\b.
Note also that this isn't perfect, but does handle the situations that you have suggested.
It's a terrible idea to begin with. You think, your users would write something like "f-aeen" to avoid your filter but would not come up with "ffaen" or "-faen" or whatever variation that you did not prepare for? This is a race you cannot win and the real loser is usability.

Need some ideas on how to acomplish this in Java (parsing strings)

Sorry I couldn't think of a better title, but thanks for reading!
My ultimate goal is to read a .java file, parse it, and pull out every identifier. Then store them all in a list. Two preconditions are there are no comments in the file, and all identifiers are composed of letters only.
Right now I can read the file, parse it by spaces, and store everything in a list. If anything in the list is a java reserved word, it is removed. Also, I remove any loose symbols that are not attached to anything (brackets and arithmetic symbols).
Now I am left with a bunch of weird strings, but at least they have no spaces in them. I know I am going to have to re-parse everything with a . delimiter in order to pull out identifiers like System.out.print, but what about strings like this example:
Logger.getLogger(MyHash.class.getName()).log(Level.SEVERE,
After re-parsing by . I will be left with more crazy strings like:
getLogger(MyHash
getName())
log(Level
SEVERE,
How am I going to be able to pull out all the identifiers while leaving out all the trash? Just keep re-parsing by every symbol that could exist in java code? That seems rather lame and time consuming. I am not even sure if it would work completely. So, can you suggest a better way of doing this?
There are several solutions that you can use, other than hacking your-own parser:
Use an existing parser, such as this one.
Use BCEL to read bytecode, which includes all fields and variables.
Hack into the compiler or run-time, using annotation processing or mirrors - I'm not sure you can find all identifiers this way, but fields and parameters for sure.
I wouldn't separate the entire file at once according to whitespace. Instead, I would scan the file letter-by-letter, saving every character in a buffer until I'm sure an identifier has been reached.
In pseudo-code:
clean buffer
for each letter l in file:
if l is '
toggle "character mode"
if l is "
toggle "string mode"
if l is a letter AND "character mode" is off AND "string mode" is off
add l to end of buffer
else
if buffer is NOT a keyword or a literal
add buffer to list of identifiers
clean buffer
Notice some lines here hide further complexity - for example, to check if the buffer is a literal you need to check for both true, false, and null.
In addition, there are more bugs in the pseudo-code - it will find identify things like the e and L parts of literals (e in floating-point literals, L in long literals) as well. I suggest adding additional "modes" to take care of them, but it's a bit tricky.
Also there are a few more things if you want to make sure it's accurate - for example you have to make sure you work with unicode. I would strongly recommend investigating the lexical structure of the language, so you won't miss anything.
EDIT:
This solution can easily be extended to deal with identifiers with numbers, as well as with comments.
Small bug above - you need to handle \" differently than ", same with \' and '.
Wow, ok. Parsing is hard -- really hard -- to do right. Rolling your own java parser is going to be incredibly difficult to do right. You'll find there are a lot of edge cases you're just not prepared for. To really do it right, and handle all the edge cases, you'll need to write a real parser. A real parser is composed of a number of things:
A lexical analyzer to break the input up into logical chunks
A grammar to determine how to interpret the aforementioned chunks
The actual "parser" which is generated from the grammar using a tool like ANTLR
A symbol table to store identifiers in
An abstract syntax tree to represent the code you've parsed
Once you have all that, you can have a real parser. Of course you could skip the abstract syntax tree, but you need pretty much everything else. That leaves you with writing about 1/3 of a compiler. If you truly want to complete this project yourself, you should see if you can find an example for ANTLR which contains a preexisting java grammar definition. That'll get you most of the way there, and then you'll need to use ANTLR to fill in your symbol table.
Alternately, you could go with the clever solutions suggested by Little Bobby Tables (awesome name, btw Bobby).

Categories