Sign in to see all reviews and comparisons. It's Free!
By clicking Sign In with Social Media, you agree to let PAT RESEARCH store, use and/or disclose your Social Media profile and email address in accordance with the PAT RESEARCH Privacy Policy and agree to the Terms of Use.
ChatterBot is a Python library that makes it easy to generate automated responses to a user’s input.
Category
Chatbot Platform
Features
• Get input • Process input • Return response
License
Proprietary
Price
Contact for Pricing
Pricing
Subscription
Free Trial
Available
Users Size
Small (<50 employees), Medium (50 to 1000 Enterprise (>1001 employees)
Company
ChatterBot
What is best?
• Get input • Process input • Return response
PAT Rating™
Editor Rating
Aggregated User Rating
Rate Here
Ease of use
8.1
6.7
Features & Functionality
8.3
8.3
Advanced Features
8.2
9.5
Integration
8.3
—
Performance
8.1
4.4
Training
10
Customer Support
8.2
4.4
Implementation
—
Renew & Recommend
—
Bottom Line
Chatterbot is a machine learning and conversational dialog engine.
8.2
Editor Rating
7.5
Aggregated User Rating
6 ratings
You have rated this
Chatterbot is a machine learning and conversational dialog engine. Chatterbot enables easy generation of automated responses to the use’s input. In order to generate different types of responses, Chatterbot employs use of machine learning algorithms therefore enabling developers automate conversations and create chat bots for users in an easy way. Chatterbot is language independent.
The language independent design, enables Chatterbox speak any natural language. Due to Chatterbot’s machine language nature a human agent or developer is also able to improve its knowledge level. This improves Chatterbot’s level of relaying possible responses when it interacts with other sources of informative data as well as humans. Chatterbot works by first creating a Python library.
This is an instance where the Chatterbot is untrained. The Python library is essential as it easily enables creation of software that can engage in user’s conversation. The library will be the location where Chatterbot will be saving a text from a statement once it is entered and when a statement is delivered as a response to the user.
The level of accuracy of responses by the Chatterbot increases when more input text is fed to the Chatterbot. The input statement will be processed by two logic adapters. The Chatterbot then later selects a matching response to the closest matching statement that matches the input.
The response submitted from the Chatterbot will be from the selection of known responses related to the statement. Also the returned response from the logic adapters is chosen based on the confidence value level based on the matching response.
Very limited features and functionality. It has nothing to do with AI. It works exclusively on reset prompts and responses, something that even a simple Python script can do. And in fact, better, that is, it could say that it has no response for a question or statement that is not found in the chat database, instead of issuing totally irrelevant responses, as CB does under these circumstances.
Performance1.5
It takes very long for the CB modules to load and way too long to train a DB, esp. using the 'UbuntuCorpusTrainer' with TSV files. E.g. it takes 2.5 min for 94 TSV files from the 'ubuntu_dialogs' corpus. To train the whole 'ubuntu_dialogs\9' folder, I have calculated that it would take 1 hour and 8 minutes! But this would not consist an issue --since this is done only once-- if CB could exploit well the 500MB (!) database that such a training would create and not behave in a totally stupid manner, as I explained in the features & functionality section. So, I can say that it's performance in general is pitiful.
Training 5
Easy, but very slow to extremely slow.
Implementation1.5
(See PERFORMANCE section)
ADDITIONAL INFORMATION I have no additional information to give.
ChatterBot fails totally
It is certainly easy to use
Very limited features and functionality. It has nothing to do with AI. It works exclusively on reset prompts and responses, something that even a simple Python script can do. And in fact, better, that is, it could say that it has no response for a question or statement that is not found in the chat database, instead of issuing totally irrelevant responses, as CB does under these circumstances.
It takes very long for the CB modules to load and way too long to train a DB, esp. using the 'UbuntuCorpusTrainer' with TSV files. E.g. it takes 2.5 min for 94 TSV files from the 'ubuntu_dialogs' corpus. To train the whole 'ubuntu_dialogs\9' folder, I have calculated that it would take 1 hour and 8 minutes! But this would not consist an issue --since this is done only once-- if CB could exploit well the 500MB (!) database that such a training would create and not behave in a totally stupid manner, as I explained in the features & functionality section. So, I can say that it's performance in general is pitiful.
Easy, but very slow to extremely slow.
(See PERFORMANCE section)
ADDITIONAL INFORMATION
I have no additional information to give.