SpellCheck and other Cleaning Techniques

Just a reminder as we head into the Holiday season that we will not be doing blog posts for the next 2 weeks as you are probably going to be too busy anyways.  Have a Merry Christmas and a Happy New Year and we will be back every Monday in the 2019

If you work a lot with data you have probably heard a lot about data cleansing. There are various ways to perform this. For CRM there are ways to split full names into first, middle and last names. There are ways to validate postal codes. The same applies to AI and ML you need to help the provider with a good data set.

Before doing Natural Language Processing there are 3 general things you can do to improve the success of your processing:
  1. Do language detection. This will help because if the text is in French then all your English models for NLP will fail. 
  2. Run SpellCheck on your text. Although this seems simple it is often missed when doing NLP. If you use a very basic dictionary then it will catch any words that are not common words in that langauge.

    Use the spelling errors to help determine if there is a topic specific model that should be used. For example if SpellCheck returned Sitecore as a spelling error then you know the article or tweet is about Sitecore and can use that model for more success.

    Sometimes spelling errors are on purpose for example Mark Stiles (a Cognitive Services Guru) has a twitter handle of @maaakstiles so sometimes if people are tweeting to him they may do similar to other words like aaaaiiii is awesome.  There may be that running joke about things being Legen ... wait for it ... dary. Those are things that a spell check would find and you can flag them and then run your NLP with it corrected but keep the context.
  3. Run a word counter or popular word finder. There may be words in the dictionary that are still special. The post popular words could help you determine context and as mentioned in the previous article context will help improve your NLP
With these 2 things you can dramatically improve the success of your Natural Language Processor.



If you are interested in learning more then reach out and let's discuss. I will be providing a basic QuickStart to SitecoreDain subscribers soon. If you cannot wait email me at chris.williams@readwatchcreate.com and as a subscriber, I can release an early version to get you started now.


Comments

Post a Comment

Popular posts from this blog

Natural Language Processing

Another key concept to DAIN and DIANA is Natural Language Processing. There are a variety of libraries and algorithms that can convert a sentence or series of statements into their various figures of speech.  DAIN and DIANA use the provider model the same way we did for Language Detection. This allows you to swap out your NLP as well as use multiple with the same idea of providing weighting to determine which provider would be considered more important. Weighting can be a simple number weighting or could include simple or more complex rules like this provider works better for English than it does for french or if this sentence contains these keywords then use this provider over that one. Natural Language Processing Steps There are various phases in Natural Language Processing. This article explains the basic ones.  Some libraries will expose the various steps while others will group them together or have a single method to do all of them: Splitting: This is the proces...
Read more

Providers and Pipeline, Oh Why!!!

I have mentioned to a few people that DAIN and DIANA are built with providers and pipelines. The pipelines. Some providers are weighted while others run conditionally based on rules.  The question I get is often is why not just choose an algorithm and stick with it for simplicity.  There are a few reasons for that: There are 3 reasons to have different providers the first is Bias and the second is Context. Bias In the case of Natural Language Processing, you have heard a lot in the news about bias in AI and ML. Some of that is due to data which is the subject of another article but other times it is that the algorithm has been trained with data that is not specific to your domain and thus can produce bias that is not specific to your data. That is why it is so important to to use multiple algorithms to keep the other in check. You can weight one algorithm higher so it is most likely to produce the result but there are cases where it differs greatly from other algorithms t...
Read more

The Universal Content Processing Engine

I am starting to learn more about Digital Asset Management and Content Processing and automatically processing content to gather Metadata both structured and unstructured.  Through this I have come up with a new self-learning engine called the Universal Content Processing Engine. Whereas currently I take advantage of statically programmed logic on how to process a tweet, make calls to given API based on static pipelines, I am limited to how I am coded to a certain extent although some learning has been happening.  Through the use of UCP Engine, I will start with a core set of knowledge and will build upon that.  I will then ask my maker for more information on how to process different types of data. Sometimes it would be a configuration of an existing component and other times it will be another provider dynamically added to my system. To start I will process content from 3 sources: FILE, STRING, URL. Through supervised learning I will learn about more sources and ...
Read more