MonkeyLearn
MonkeyLearn is a no-code text analytics platform designed to simplify the process of obtaining, analyzing, and visualizing customer feedback. It leverages artif...
Last verified:
What is MonkeyLearn?
MonkeyLearn is a no-code, AI-powered text analysis platform that enables users to classify and extract information from unstructured text using machine learning. The platform turns raw text from emails, support tickets, social media posts, surveys, reviews, documents, and tweets into actionable, structured data without requiring any machine learning expertise or coding skills.
Key features include pre-trained models for sentiment analysis, topic detection, keyword extraction, and intent classification that can be deployed instantly. Users can also create custom classifiers to group text into defined categories (by sentiment, emotion, topic) and custom extractors to retrieve specific information like keywords, entities, phrases, and numbers. The platform offers a graphical user interface for training custom models, a scalable cloud computing platform for instant model training and execution, and APIs with SDKs for Python, Ruby, Node, Java, and PHP for integration into any software project.
MonkeyLearn is designed for small and medium companies, marketers, salespeople, customer support teams, data analysts, and developers who need to analyze large amounts of text data. It integrates seamlessly with Google Sheets, Zapier, Zendesk, Make, RapidMiner, and hundreds of other applications through direct integrations or open API, saving hours of manual text processing and enabling data-driven business decisions.
MonkeyLearn pricing
Pricing model: Free
MonkeyLearn offers a free tier with 300 queries per month that requires no credit card. Paid plans include Standard at $299 per month (monthly payment), Team at $299/month, Business at $599/month, and Enterprise with custom pricing. The pricing type is flat rate with quote-based options available. All plans include access to basic text analysis features, pre-trained models, and integrations. Additional add-ons and one-off options may be available.
MonkeyLearn pros
- No-code interface requires no machine learning expertise
- Pre-trained models available for sentiment analysis, topic detection, and keyword extraction
- Custom model builder allows training on your own data and criteria
- Scalable cloud platform trains and runs models instantly without installing software
- API and SDKs support Python, Ruby, Node, Java, and PHP integration
- Direct integrations with Google Sheets, Zapier, Zendesk, Make, and RapidMiner
- Upload CSV or Excel files for manual text processing
- Automatic text processing through API or integrations
- Model accuracy statistics provided during training
- Two model types: Classifiers for categorization and Extractors for data retrieval
- Pre-made topic labels organized in thematic groups for product feedback
- Keyword mapping feature defines labels with high-precision keywords
- Batch requests improve performance and reduce costs
- Free tier available with 300 queries per month
- Professional support for obtaining actionable data from rough texts
- Saves hours of manual data processing in tools you already use
MonkeyLearn cons
- Pricing starts at $299 per month which is expensive for small businesses
- Free tier limited to only 300 queries per month
- No credit card required for free tier but paid plans require payment setup
- Sentiment analysis has fixed labels (Positive, Negative, Neutral) that cannot be customized
- Acquired by Medallia in 2022, potential changes to standalone product direction
- Requires API token authentication for all API requests
- Custom model training requires tagging sufficient text data for accuracy
- Generic keywords can generate noise and confusion in models
Frequently asked questions about MonkeyLearn
What is MonkeyLearn and what does it do?
MonkeyLearn is a Machine Learning platform for Text Analysis that allows users to easily get actionable data from raw text. You can detect topic or sentiment expressed in texts like tweets, chats, reviews, articles, and more. It provides a graphical user interface for creating customized machine learning models, publicly available pre-trained models, a scalable cloud computing platform, and APIs with SDKs for integration.
What are the two types of models in MonkeyLearn?
Models in MonkeyLearn are organized into two families: Classification models (classifiers) that take text and return labels or categories, and Extraction models (extractors) that extract particular data within a text. Classifiers are used to group or tag data into defined categories by sentiment, emotion, topic, etc., while Extractors retrieve pieces of information like keywords, entities, phrases, and numbers.
Can I create custom models without coding experience?
Yes, anyone can build a custom model with MonkeyLearn without technical expertise or coding. The platform walks users through a wizard where they can define categories and tag text data that will train their custom model. When enough text data is tagged, MonkeyLearn offers statistics on the model's accuracy.
How do I get started with MonkeyLearn?
To start, visit monkeylearn.com and click the sign-up button to create your account. You can sign up for the free tier without entering a credit card, which gives access to basic features. Once in the dashboard, explore the tutorial or onboarding flow, try pre-built sentiment analysis, topic, and keyword extraction models, and check the limits of your tier to know when upgrading makes sense.
What integrations does MonkeyLearn support?
MonkeyLearn can be easily integrated with Google Sheets, Zapier, Zendesk, Make, and RapidMiner without coding required. It also offers an open API and beautiful SDKs for Python, Ruby, Node, Java, and PHP, allowing integration with hundreds of other applications and any software project using any programming language.
How do I process text files in MonkeyLearn?
Text can be processed manually by uploading a CSV or Excel file to your models. MonkeyLearn will process the file and present it for download with the results included. Automatic processing can be done through the MonkeyLearn API or by using Zapier, Google Sheets Extension, or RapidMiner pipelines.
What pre-trained models are available?
MonkeyLearn provides publicly available pre-trained models for common problems including sentiment analysis, topic detection, keyword extraction, and intent classification. These models can be deployed instantly on any text data. A number of pre-trained classifiers and extractors are already public and available for users without needing to train custom models.
How do I define labels for custom topic models?
Each topic label must be defined by mapping keywords to it. Select the label in the label list, check if keywords are already mapped in the Mapped Keywords list, and map keywords by clicking the arrow button from the Unused Keywords list. You can add custom keywords by writing them in the text box and clicking Add. Focus on specific, high-precision keywords that only fit one particular label to avoid noise.
What is the sentiment analysis model's label structure?
The sentiment enrichment comes with a fixed list of labels that are non-customizable: Positive, Negative, and Neutral. When adding a sentiment model, there's no need to add labels because a pre-made model for sentiment will be used automatically.
How does MonkeyLearn handle keyword extraction?
The Keywords model extracts the most relevant keywords within a text, including aspects (typically nominal groups) like 'food', qualifiers (typically adjectives) like 'delicious', and opinions (a combination of aspect and qualifier) like 'delicious food'. This extraction model identifies the most pertinent terms that represent the main subjects and sentiments in the text.