NLP vs. Machine Learning Approaches in Chatbot Instances

Chatbot instances can act very differently depending on whether they're based on NLP or machine learning approaches. Here's a practical example of that.

What is NLP?

NLP or Natural Language Processing is a sub-field of AI that aims to teach machines how to understand human languages.

Our Co-founder and CEO, Jordi Torras, explains it better than anyone. 

Today, industry-leading NLP is built on AI that detects patterns in data, together with linguistics models and accumulated lexical, and semantic knowledge which make it possible to understand user inputs. This results in an approach that is flexible in adapting to the nuances and ambiguity of languages while boosting accuracy between results and user intents.

This means NLP is semantically sensitive, and does not base its answers on the exact input keywords, but instead focuses on the real content and real meaning behind queries.

Because it is a rule-based reasoning system, NLP also provides easy visualization of the reasoning behind every answer. If at any point the chatbot provides a wrong answer, the chatbot project team can trace the problem, discover where it went wrong, and fix the issue.

How is Machine Learning different from NLP?

In the past few years, Machine Learning has become a trendy keyword. Practically all companies on the face of the earth seek to implement some sort of machine learning-based tool, sometimes for the sake of coolness. But is it really worth it? Well, it depends on the field of application.

Machine Learning is a computer science field that provides systems the ability to automatically learn and improve from past behaviors based on statistics. This requires large amounts of information from which to extract patterns that will serve to deliver results. 

When it comes to search and chatbots, what this really means is that there is no real understanding of the human language involved. This approach can work for companies like Google, which have tons and tons of user data to base their patterns on, but with smaller samples of data, the results are likely to be limited in terms of accuracy. What usually happens is that machine learning-based chatbots are not predictable and can come up with deformed patterns that truly affect the performance and quality of your chatbot answers.

Plus, whenever there is an issue, developers won’t be able to quickly fix it, as machine learning uses a sort of “black box” where decisions happen that is not really understandable or modifiable by developers. In order to correct the issues, a lot of training will be needed, and performance might be damaged even more on the way.

Controlling the quality of your chatbot conversations requires therefore an NLP-based approach, in order to ensure the right answers are provided.

NLP vs Machine Learning: we put our NLP to test

In a short comparative exercise, we put another NLP-driven chatbot against Inbenta’s own to examine their real responses to “natural” language. The following questions were asked both to the Inbenta Chatbot and to another popular chatbot service on the market that advertises its use of NLP. For the sake of this comparison, we’ll refer to it as the ‘anonyBot’.

Without making any changes to our lexicon, we took the FAQs used by anonyBot and plugged them into our Inbenta knowledge base.

The results of the experiment pointed out a clear discrepancy between the two conceptions of Natural Language Processing.

To further demonstrate one of the many steps in our NLP process, we’ve tagged each result shown here with its semantic score — a calculated percentage of how close of a match the ending result is to the user’s original query.

How semantic scoring works:

Using Inbenta’s NLP stack, we’re able to abstract a query down to clusters of lexical units, functions, and concepts that represent meaning, or intent, rather than language. We then use an algorithm that ranks existing contents to the intent, and the strongest match is returned to the customer in the chatbot’s response.

NLP vs Machine Learning – The experiment:

Question 1: Is there a limit of the size of db?

anonyBot: No answer [✗]

Inbenta: How large a knowledge base can I create? [✓]

Our semantic score: 58.8%

Question 2: Which formats do you support?

anonyBot: No answer [✗]

Inbenta: What format does the tool expect the file content to be? [✓]

Our semantic score: 80.9%

Question 3: I forgot my password

anonyBot: No answer [✗]

Inbenta: How do I login? [✓]

Our semantic score: 57.1%

Question 4: I made a change but I can’t see it yet

anonyBot: No answer [✗]

Inbenta: The updates I made to my knowledge base are not reflected on publish. Why not? [✓]

Our semantic score: 58.4%

Question 5: Is my data secure?

anonyBot: No answer [✗]

Inbenta: How safe is my knowledge base data? [✓]

Our semantic score: 92.4%

Infographics on NLP vs Machine Learning

Want to know more about NLP and machine learning? Check our infographics.

Inbenta is a leader in natural language processing and artificial intelligence for customer support, e-commerce and conversational chatbots, providing an easy-to-deploy solution that improves customer satisfaction, reduces support costs, and increases revenue.

Interested in finding out more? Our team of experts is available to show you how Inbenta can benefit your company

Check out our similar articles