Data Mining: Text Mining, Visualization and Social Media

Google Calendar

I went ahead and tested Google Calendar (finally, it came up) against the free text input I threw at SpongeCell:

• Visiting Granny at 12 on March 1st: passed - SpongCell passed.
• Flying to Cambridge on the 2nd then back on the 4th: failed - SpongCell passed.
• Flying to Boston tomorrow: passed - SpongeCell passed.
• Flying to Boston in a week: failed - SpongeCell failed.
• Flying to Boston on Thursday: passed - SpongeCell passed.
• Flying to Boston a week on Thursday: passed - SpongeCell failed.
• Flying to Boston on the 30th of Feb: same as SpongeCell - placed as 2nd March.

Overall, it seems that SpongeCell is ahead of Google in terms of NL interfaces. It will be interesting to revisit these sites in a while with a new set of tests to see how they have improved.

Online Calendars and NLP Interfaces

Paul Kedrosky writes about Google Calendar:

The highlights: Gmail integration (in principle), especially the recognition of event info in email content; the spiffy interface; the atomized calendar sharing; and the outrageously cool English syntax for adding events ("Dinner with Steve tomorrow at 7pm" does what you think it should).

I've not yet used it myself, but it is interesting to note that that the online calendar space seems to be one area where natural language interfaces seems to be taking off - c.f. my post on SpongeCell.

While writing the above I thought - why not test Google Calendar with the phrases I threw at SpongeCell. Logging on to GC I find:

Calendar is unavailable right now, please try again in a few moments

SpongeCell

TechCrunch writes about a new calendaring service: SpongeCell. The description mentions the use of NLP to grok free text entries to the caledar, so I had to give it a shot:

• Visiting Granny at 12 on March 1st: no problem, off to a good start.
• Flying to Cambridge on the 2nd then back on the 4th: sweet - it got the two dates. They were placed in February, so no context (my last entry was March), but that is probably the right behaviour.
• Flying to Boston tomorrow: this got entered in today's field, could be a time zone problem. It is 2am on the 31st where I am, so the entry should have been on the 1st of February.
• Flying to Boston in a week: nope, turned up in yesterday's list. Could be related to the issue above in that it really meant to put it in today's list - either way it's wrong.
• Flying to Boston on Thursday: no problem.
• Flying to Boston a week on Thursday: nope - just Thursday.
• Flying to Boston on the 30th of Feb: oops - turns up on the 2nd of March. An understandable error, but certainly a corner case that needs to be addressed.

In summary, the NLP does a reasonable job. However, there is always a danger with this type of interface that the user will learn what can and can't be entered and then restrict their input to a closed set of patterns - essentially a controlled language for date entry. This doesn't actually undermine the utility of the input method (think of the Graffiti method to input text into a Palm device).