Explore Courses Blog Tutorials Interview Questions
+1 vote
in Machine Learning by (3.5k points)
edited by
I would love to know in iOS email client, How apple manages to create an appointment or set a destination in map by just a single click on the hyperlink created automatically in emails which contains date, time or a place? It would be really interesting to know what is the processing behind this even in the case of languages other than English?

I think they use some machine learning or run a lot of regular expression at ones. But I have a doubt that What they do to make these stuff  work on multiple languages or date time format.

So, what do you think which machine learning algorithms they are using? Can someone tell me how do they manage to extract entities in a very short time to accomplish this task?

And how do they manage to do this for multiple languages this quick? Which machine learning algorithm they are using to do this task?

2 Answers

0 votes
by (46k points)

the technology apple uses is Apple Data Detectors  which parses the text and represent piece of data by detecting the pattern. To know more click here.

+4 votes
by (10.9k points)
edited by

@Ritik, Apple has a technology of their own called the Apple Data Detectors.It parses the text first and then detects the patterns having specific pieces of data and then applies Os-contextual actions to it.

Alternatively, the same thing can be done using the information extraction technique which automatically extracts structured info from unstructured machine-readable documents.You may use Stanford's SUTime tool which allows you to extract attributes about n-grams in a document like:

1. numberOfSymbols

2. Length

3. nextWord

4. previousWord

5. numberOfLetters

 Then it uses the classification algorithm and feeds both negative and positive examples.

The algorithms learned based on those examples can be applied to future examples which it hasn’t seen before.


Observed     Letter  Symbols   length     PWord   NWord     Date  

"MAR."             3           1               4         "Tue"     "25th"     TRUE  

"APR"              3            0              3       "school"   "while"     FALSE  

Through this examples it may learn rules like:

  • If the previous words (PWord) is only characters.
  • The current word is in “March”.
  • The next word (NWord) is in which number.
  • Then the date

Browse Categories