An Analysis of the Best NLP Tool to Build a Conversational Bot

‘Chatbots’ — a term which is familiar for the layman. We could see and experience the usage of chatbot in our daily life. With the advent of technology, changes in consumer’s behavior and expectation also increases.Through technological expansion, chatbots and intelligent assistants are redefining the user engagement and brand experiences.

It’s better to start with a quote from Ted Livingston, founder, and CEO of Kik: “Chat apps will come to be thought of as the new browsers; bots will be the new websites. This is the beginning of a new internet.

What are chatbots?

In simple words, chatbot is a software chunk that helps to chat with things to get done or to be entertained. It is a technology which aids the human to converse with the computers in their own language using a graphical interface. Chatbots tend to mimic humans in performing any tasks or conversing.

With the technological revolution, chatbots evolve in 1950 with Alan Turing’s intelligent machine.Many artificial intelligent systems emerge in the course of time following Turing’s intelligent machine.

Importance of Chatbots

Chatbots are gaining popularity in the implementation of real-world applications. They can cater the customer experience by providing customer support & service, product recommendations & suggestions and engaging customers with appealing marketing campaigns.Chatbots can function as a lawyer, teacher, doctor, personal stylist, toy and even as a companion thus enriching our life.

Building a Chatbot

A successful chatbot behaves to be human-like and it provides valuable features that help users. For building a chatbot we have to identify its function & purpose and then corresponding sector for which we are using the same.

Tools for building Chatbot

We got a basic idea about the chatbots and its importance.Now the question arises ‘Where and How could I build these chatbots?’There are several tools available in the market for building chatbots.The tools are Dialogflow, wit.ai, Amazon lex, Luis, Watson and so on.But still, a confusion occurs ‘Which tool should be used for these?’The following section discusses these tools:

1. DialogFlow (Api.ai)

Api.ai is now known as DialogFlow. It was acquired by Google in 2016. This platform connect with users on Google assistant and other messaging apps and devices.Here is an excerpt from the CEO of DialogFlow at ProductHunt about the characteristic features of DialogFlow.

DialogFlow uses Intents, Entities, Actions with parameters, Contexts, Speech to Text and Text to Speech capabilities, along with machine learning that works silently and trains your model. DialogFlow has built-in knowledge on topics like casual talks, weather, and wisdom. It means we don’t have to train the agent for these intents. DialogFlow returns the output as JSON data.

• Programming Languages / Apps /Integration : SDKs — Android, iOS, Cordova, JavaScript, HTML, Node.js, .NET, Unity, Xamarin, C++, Python, Ruby, PHP (Community Supported ), Epson Moverio, Botkit, JAVA. Built in integration with Google Assistant, Facebook Messenger and Slack.
• Platform : Google Assistant and other messaging platforms
• Supported Languages : Brazilian Portuguese, Chinese (Cantonese), Chinese (Simplified), Chinese (Traditional), English, Dutch, French, German, Italian, Japanese, Korean, Portuguese, Russian, Spanish, Ukrainian.
• Pricing : Free
• Limit for API calls : Unlimited
• Third Party Integration : Allowed for different messengers and some additional services.
• Suitable for : Middle level B2C bots, Virtual Assistants, MVPs

2. Wit.ai

Wit.ai is acquired by Facebook. It is a SaaS platform making easy for developers to build a chatbot for their app or device. It can be used in platforms which learn new commands semantically to those input by the developer.

• Programming Languages / Apps / Integrations : Node.js client, Python client, Ruby client, On other platforms: HTTP API
• Platform : Applications and devices that you can talk or text to.
• Supported Languages : Albanian, Arabic, Azerbaijani, Bengali, Bosnian, Bulgarian, Burmese, Catalan, Chinese, Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, Georgian, German, Greek, Hebrew, Hindi, Hungarian, Icelandic, Indonesian, Italian, Japanese, Korean, Latin, Lithuanian, Macedonian, Malay, Norwegian, Persian, Polish, Portuguese, Romanian, Russian, Serbian, Slovak, Slovenian, Spanish, Swahili, Swedish, Tagalog, Tamil, Thai, Turkish, Ukrainian and Vietnamese.
• Pricing : Free
• Limit for API calls : Unlimited
• Third Party Integration : Doesn’t support third-party integration tools.
• Suitable for : Simple B2C bots, MVPs

3. Amazon lex

Amazon lex is a service for building chatbots for any application or device using voice and text. Using Amazon lex any developer can build conversational chatbots instantly.It is the service provided by Amazon. Amazon Lex provides the advanced deep learning functionalities of automatic speech recognition (ASR) for converting speech to text, and natural language understanding (NLU) to recognize the intent of the text.

• Programming Languages / Apps / Integrations : IoS and Android SDKs, Java, JavaScript, Python, CLI, .Net, Ruby, PHP, Go, and C++
• Platform : Integration with Facebook, Slack and Twilio; but if you want to connect to a messaging device / application which isn’t supported, there’s an API for that — the Amazon Lex API.
• Supported Languages : Amazon Lex is limited to support of US English.
• Pricing : Trial — 1 year ; Paid — $0.004 per speech query & $0.00075 per text query
• Limit for API calls : Trial — 10k speech queries, 5k text queries ; Paid — Unlimited
• Third Party Integration : Different messengers, Amazon services, popular SaaS platforms.
• Suitable for : Preview mode, just for making an opinion

4. Luis

Luis is acronym for Language Understanding Intelligent Service. It is the service provided by Microsoft. Luis understand the language of laymen contextually and communicate in the way they speak. It also uses intents and entities. One of the key feature of Luis is active learning technology. It offers pre-built domains for things like devices, music, calendar and so on.

• Programming Languages / Apps / Integrations : C# SDK, Python SDK, Node JS SDK, Android SDK.
• Platform : Facebook, Kik, Slack, Telegram, Twilio, Microsoft Teams, Skype,k GroupMe, Web Chat, Email, Direct Line.
• Supported Languages : English, French, Italian, German, Spanish, Brazilian Portuguese, Japanese, Korean and Chinese.
• Pricing : LUIS API — Free : 10,000 transactions free per month; LUIS API — Basic : Up to 10 transactions per second; $0.75 per 1,000 transactions.
• Limit for API calls : Free — 10,000 queries per month & 5 queries per second. Paid — 10 queries per second; $0,75 per 1,000 queries.
• Third Party Integration : Microsoft Azure, different messengers via bot framework.
• Suitable for : Cortana functionality, IOT applications and virtual assistant apps, bots integrated to enterprise software.

5. IBM Watson

Watson is an AI platform provided by IBM. It can understand all forms of data, interact with people and learn from that interaction. IBM had moved the technology of Watson into cloud and released it API allowing the user to make conversational bots. It is built on neural network. The main components are intents, entities and dialog.

• Programming Languages / Apps / Integrations : Node SDK, Java SDK, Python SDK, iOS SDK, Unity SDK
• Platform : Any messaging platform supporting speech, images, and text.
• Supported Languages : English and Japanese.
• Pricing : Lite, Standard & Premium plans are available for Watson. Lite — 1,000 queries per month ; Standard — $0.0025 per API call ; Premium — Price as per the request.
• Limit for API calls : Lite — 1,000 API queries per month ; Standard — Unlimited API queries per month , Upto 20 workspaces , Upto 2,000 intents ; Premium — Unlimited.
• Third Party Integration : IBM Services
• Suitable for : Suitable for virtual assistants and bot that integrate with IBM services.

Popularity of NLP Tools

We are witnessing a huge demand for the growth of chatbots. Currently, a large number of platforms are available for making chatbots. The tools discussed above are backed by Tech — Giants. So we could say they all maintain a standard. The below figure shows the google trends in the conversational interfaces.

Wrapping Up

In order to create a chatbot for your business, a variety of NLP tools are available. The choice of tool depends on the requirements & needs, integration needed and technology preference. There are much more platforms to explore. We had given a glimpse of few frameworks or tools that would be useful for building chatbots. With the pace of progress and interest in conversational interfaces, new tools may soon arise in the market.

Want to explore more about conversational interfaces we are here to help you!

Contact Us Today!

Originally published in Cabot Solutions Blog on December 29, 2017