After seeing the disadvantages of not using Natural Language Processing, you are now going to explore how a NLP improved search feature can greatly improve the user experience of your data-driven apps. In fact, many users are going to expect or are already expecting more advanced search because they are used to it from apps they are using day by day. Notes, reminders, or the Photos app are already NLP enhanced.
- So here again, we have our diary application but this time, it's version 2.0 and it is NLP enhanced and what this means, I am going to show you right now. So if we're entering buy again, what we already did before, we're getting a lot more search results and this is because we have new NLP capabilities here which also gives us results when we have another form, or inflected form of our word, buy. So we're also finding, this is a great day.
I bought my first iPhone. And here we have the inflected form of buy which is bought. And another example would be someone told me that all cats have green eyes. I'm just not buying it. So we're also getting this result although we searched just for the word buy. And this doesn't only work in english. It also works in other languages and we're even detecting other languages here. So, if I, for example, enter kaufen, which is the german verb for to buy, we're getting, Wir kaufen uns bald einen Homepod, which means we're going to buy a Homepod soon.
And we also get, ich habe mir ein neues iPhone gekauft, which means, I just bought a new iPhone. And here we have a different form of the verb buy, which is now gekauft. And the original form is kaufen. So this process is called lemmatization, it works in different languages so we can identify the language, we can use lemmatization. And since this is such an important concept in natural language processing, you're going to hear a little bit more about that soon. And, one very important aspect of this is, that these search behaviors, finding different verbs in different forms is expected by the users because they know nothing else from other applications like Notes, for example.
So it is important that when you implement such a search for your own application, that you also take into account the power of natural language processing and the benefits for your users.
- What are machine learning, Core ML, Vision, and NLP?
- Adding a machine learning model to a project
- Getting predictions from machine learning models
- Converting existing machine learning models for Core ML
- Classifying images and detecting objects with Vision and Core ML
- Analyzing natural language text with NSLinguisticTagger