Workflows Available in Relevance AI
Relevance AI offers a variety of very useful analyses that are all accessible under Workflows on the navigation bar. Our workflows are prepared in a no-code format, meaning you do not need any programming skills to be able to use them. Alternatively, if you are into Python programming, you can use our Python SDK.
Below are the top workflows available in Relevance AI.
All Data Types
Vectorize
Vectorize your data with a ready-to-use neural network, and benefit from the magic of AI and machine learning.
| Benefits |
|---|
| 1. More precise data analysis 2. Fast processing of large amounts of data 3. Easy access to a variety of state-of-the-art vectorizing models |
| Top Use Cases |
|---|
| Semantic search Clustering Recommendation system |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI 2. Choose the model and the fields you want to encode 3. Run Vectorize workflow |
Clustering
Cluster
Group/cluster your data based on the content similarity, and discover hidden patterns using AI and machine learning.
| Benefits |
|---|
| 1. Identify hidden patterns in your data using AI algorithms 2. Understand your data better 3. Fast and more informed decision making |
| Top Use Cases |
|---|
| Analysing top trends in reviews Understanding different categories in costumer feedback (issues, requests, admiration) |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Vectorize the field in your data to be analyzed 3. Run Clustering workflow |
Cluster Auto
Group/cluster your data based on the content similarity, and discover hidden patterns using AI and machine learning. The number of groups/categories is decided automatically.
| Benefit |
|---|
| 1. Identify hidden patterns in your data using AI algorithms 2. Understand your data better 3. Fast and more informed decision making 4. Does not require you to specify the number of clusters in advance |
| Top Use Cases |
|---|
| Analysing top trends in reviews Understanding different categories in costumer feedback (issues, requests, admiration) When there is no good estimate of the number of clusters |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Vectorize the field in your data to be analyzed 3. Run Auto-clustering workflow |
Cluster Hybrid
Hybrid-clustering workflow combined several clustering methods for grouping data.
| Benefit |
|---|
| 1. Identify hidden patterns in your data using AI algorithms 2. Understand your data better 3. Fast and more informed decision making 4. Combines different clustering methods |
| Top Use Cases |
|---|
| Analysing top trends in reviews Understanding different categories in costumer feedback (issues, requests, admiration) When there is no good estimate of the number of clusters |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Vectorize the field in your data to be analyzed 3. Run Hybrid-clustering workflow |
Subcluster
Subcluster existing clusters/categories into smaller ones for a higher level of granularity.
| Benefits |
|---|
| 1. More fine-grained analysis of the data 2. Extract all the hidden information about your data |
| Top Use Cases |
|---|
| Understanding the inner categories of a category (e.g. for a category focused on water leakage, subcategories could be "leakage in the kitchen" or "leakage in the balcony") |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Vectorize the field in your data to be analyzed 3. Run clustering workflow 4. Run Subcluster workflow |
Classify Vectors on Existing Clusters
| Benefits |
|---|
| Coming Soon |
| Top Use Cases |
|---|
| Coming Soon |
| Start Analyzing |
|---|
| Coming Soon |
Reduce Dimensions
Reduce the size of vectors associated with data entries.
| Benefits |
|---|
| 1. Reduce the amount of time and memory needed to work with vector analysis 2. Possibility to explore data in a 3D space |
| Top Use Cases |
|---|
| 2D or 3D data representation Increase the efficiency of using resources (memory, computation) |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Vectorize the field in your data to be used for analysis or visualization 3. Run dimensionality reduction workflow on the vectors |
Text Data
AI Clustering
AI Clustering is an advanced workflow; your one-stop-shop to transform text data to insight. This workflow categorize your data and present the results on the Explorer dashboard. There are optional steps such as sentiment and emotion analysis that you can add to the process when using the setup wizard.
| Benefits |
|---|
| 1. Topic modelling with a few clicks 2. Access to other processing such as emotion and sentiment analysis at the same time 3. Easy representation of the data |
| Top Use Cases |
|---|
| 1. Customer feedback analysis 2. Product review analysis |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run AI Clustering workflow |
AI Tagging
AI Tagging is an advanced workflow composed of several processing steps. This workflow applies tags (i.e. code-frames) to your data. The tags are generated by strong AI models. Through the workflow, you can modify the extracted tag list using your domain knowledge. Finally review the results and build up your explorer app.
| Benefits |
|---|
| 1. Topic modelling with a few clicks 2. Access to a user-friendly review tool 3. Easy representation of the data |
| Top Use Cases |
|---|
| 1. Customer feedback analysis 2. Product review analysis |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run AI Tagging workflow |
Cluster your text data
| Benefits |
|---|
| Coming Soon |
| Top Use Cases |
|---|
| Coming Soon |
| Start Analyzing |
|---|
| Coming Soon |
Cluster Text (One-To-Many)
| Benefits |
|---|
| 1. Topic modelling with a few clicks 2.Identifying various topics included in a single piece of text |
| Top Use Cases |
|---|
| 1. Customer feedback analysis 2. Product review analysis |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run Cluster Text (One-to_Many) |
Extract Sentiment
Perform sentiment analysis on your textual data using state-of-the-art AI and machine learning to discover text polarity (i.e. positive, negative, neutral).
| Benefits |
|---|
| 1. Understand your customer better 2. Identify areas of strong/weak points |
| Top Use Cases |
|---|
| Understanding clients better Understanding how staff feel about different aspects of work |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run sentiment analysis workflow |
Extract Emotion
Perform emotion analysis on your textual data using state-of-the-art AI and machine learning to discover different underlying emotions in your data (i.e. anger, satisfaction, happiness, disapproval, etc.)
| Benefits |
|---|
| 1. Understand your customer better 2. Identify areas of strong/weak points 3. More fine-grained analysis compared to sentiment identification |
| Top Use Cases |
|---|
| Understanding clients better Understanding how staff feel about different aspects of work |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run emotion analysis workflow |
Split to Sentences
Use AI to split large chunks of texts to their composing sentences.
| Benefit |
|---|
| The ability to analyse all points mentioned in comments and reviews separately |
| Top Use Cases |
|---|
| Understand the comments better fine grained analysis of the action points |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run the split to sentence workflow |
Tagging
Generate AI Tags
Use AI to automatically generate AI tags. This includes tag/code frame extraction and labelling the dataset with the tags.
| Benefits |
|---|
| 1. Use AI to automatically generate conceptual keywords/phrases (code frames) for your data 2. Use the code frames to better understand your data 3. Data categorisation using the code frames |
| Top Use Cases |
|---|
| Understand the data better by extracting the top code frames Date categorisation relying on automatically extracted and conceptual key phrases |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run the generate AI tags workflow |
Guided Tagging
Guided tagging workflow enables you to modify (add to/ remove from) previously extracted tags before tagging the a dataset.
| Benefits |
|---|
| 1. Capability of modifying the automatically extracted code frame 2.Use AI to automatically assign conceptual keywords/phrases (code frames) for your data 3. Use the code frames to better understand your data 4. Data categorisation using the code frames |
| Top Use Cases |
|---|
| Understand the data better by extracting the top code frames Date categorisation relying on automatically extracted and conceptual key phrases |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run the generate AI tags workflow 3. Run Guided tagging workflow which allows you to modify the extracted code frames if necessary and then apply them to the dataset |
Sub-Tagging
Sub tagging allows you to perform a more fine-grained analysis on a large category. Sub-tagging breaks a parent category to a defined set of child categories.
| Benefit |
|---|
| 1. More fine-grained analysis of the data 2. Controlled parent to child categorization |
| Top Use Cases |
|---|
| Analysing client feedback and when extracted categories need to be broken into their composing categories |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run a tagging workflow (e.g. generate AI tags workflow ) 3. Run Sub tagging workflow. |
Guided Sub-Tagging
Guided sub-tagging enables you to tag a field based on another tagging results. It is specially useful when you want to combine multiple tagging techniques. Or there are different text fields in a dataset where the content of those fields are different but relevant.
| Benefit |
|---|
| 1. More fine-grained analysis of the data 2. Use of multiple tagging techniques or different tagged fields 3. Controlled parent to child categorization |
| Top Use Cases |
|---|
| 1. Analysing client feedback and when extracted categories need to be broken into their composing categories 2. When there are various text field in a dataset with different but relevant content |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run a tagging workflow (e.g. generate AI tags workflow ) 3. Run Sub tagging workflow. 4. Run Guided sub-tagging |
Sentiment tagging
| Benefits |
|---|
| Coming Soon |
| Top Use Cases |
|---|
| Coming Soon |
| Start Analyzing |
|---|
| Coming Soon |
Guided Sentiment Tagging
| Benefits |
|---|
| Coming Soon |
| Top Use Cases |
|---|
| Coming Soon |
| Start Analyzing |
|---|
| Coming Soon |
Tag With Your Own Tags
Tag with your own tags is recommended when a preselected code frame (i.e. tag list) is ready. In other words when AI is not required to extract candidate tags from the dataset.
| Benefit |
|---|
| 1. Allows using a pre-selected set of tags 2. Allows using AI for tag assignment |
| Top Use Cases |
|---|
| 1. When data trends are known and the user's knowledge of the dataset is vast 2. When a limited tag list is allowed |
| Start Analyzing |
|---|
| 1. Upload your dataset 2. Run Tag with your own tags workflow |
Add Tags To Existing Tags
Add tags to existing tags allows you to update an existing tag list.
| Benefit |
|---|
| 1. No need to run tagging from scratch 2. Update the tag list based on the new entries or discoveries in the dataset |
| Top Use Cases |
|---|
| 1. When some new trends/insights are discovered in data 2. When a new subset is added to a dataset |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run a tagging workflow (e.g. generate AI tags workflow ) 3. Run Add tags to existing tags |
Rename/Combine Tags
Rename/Combine Tags allows you to rename tags that are generated by other workflows. A great functionality of this workflow is to move from different child tags to one parent tag.
| Benefit |
|---|
| 1. Easy and fast rename 2. Upgrading from child tags to a parent category |
| Top Use Cases |
|---|
| When some tags needs to be renames When larger categories are preferred in analysis of the data |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run a tagging workflow (e.g. generate AI tags workflow ) 3. Run Rename/Combine Tags workflow |
Remove Tags
There might be cases where an existing tag should be removed. Remove Tags workflow provides you with a quick scan of your dataset and removes all occurrences of a specified tag.
| Benefit |
|---|
| 1. Quickly scan the whole dataset and remove an existing tag |
| Top Use Cases |
|---|
| 1. Updating the existing tags after several analysis |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run a tagging workflow (e.g. generate AI tags workflow) 3. RunRemove Tags workflow |
Translate
Translate your data to / from English.
| Benefits |
|---|
| 1. Easy access to state-of-the-art translator model 2. Understand your non English speaking client better 3. Unify the dataset by analysing the data in the same language |
| Top Use Cases |
|---|
| Translate comments from different languages and analyse them together Understand clients using a language other than English better |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run the Translate workflow |
Extract Entities
Use machine learning to extract named entitites from an existing text field.
| Benefits |
|---|
| 1. Identify the popular entities mentioned in your dataset 2. Better understand your data and users |
| Top Use Cases |
|---|
| 1. Extraction of different types of entities in data 2. Understand the data better by focusing on the top existing entities |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run Extract Entititesworkflow |
Anonymizing text
Anonymizing text is a crucial preprocessing step when working with personal data.
| Benefits |
|---|
| 1. Fast access to AI anonymizing tools |
| Top Use Cases |
|---|
| 1. Client record analysis 2. Customer feedback analysis |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run Anonymize text workflow 3. Check the results manually for any remaining sensitive information |
Extract Nouns
Use machine learning to extract nouns from text and build a taxonomy with the results.
| Benefits |
|---|
| 1. Build a taxonomy from extracted nouns 2. Use the taxonomy to better understand your data 3. Data summarisation using the taxonomy |
| Top Use Cases |
|---|
| Understand the data better by focusing on the top existing nouns Fast data summarisation using the extracted taxonomy |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run the Extract nouns workflow |
Replace Phrase
Replace phrase workflow helps you unify field values across your dataset which is of beneficial value when applying filters to categorical fields.
| Benefits |
|---|
| 1. Quick scan of a dataset to unify values under categorical fields (e.g nsw, Nsw, New South Wales--> NSW) |
| Top Use Cases |
|---|
| Filed value unification for filtering across categorical fields |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run the Replace Phrase workflow |
Extract text count
Extract features such as number of sentences, number of words and number of characters from a selected field in your dataset.
| Benefit |
|---|
| 1. Easy analysis of free-text fields in your dataset 2. Access to a numeric value to represent data under metrics and filters |
| Top Use Cases |
|---|
| Access to statistics such as which age range provides more detailed feedback |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run the Extract Text Count workflow |
Media (Image, Audio, etc.)
Connect Media
Upload Media allows you to upload your media files (e.g. images or audio files) to Relevance AI, so that the file can be processed.
| Benefit |
|---|
| 1. Easy upload 2. Free space management 3. Safe platform |
| Top Use Cases |
|---|
| 1. Image processing 2. Audio processing |
| Start Analyzing |
|---|
| 1. Create an empty dataset 2. Zip your media files (Optional) 3. Run the Upload Media workflow |
Audio Intelligence
| Benefit |
|---|
| 1. Audio Transcription 2. Speaker Diarization 3. Sentence/Chapter splitting |
| Top Use Cases |
|---|
| 1. Analysing Calls 2. Focus Groups 3. Interview analysis |
| Start Analyzing |
|---|
| 1. Create an empty dataset 2. Zip your audio files (Optional) 3. Run the Upload Media workflow 4. Run Audio intelligence workflow |
Dates
Count Number of Days
Count numebr of days workflow provides you with access to date processing tools to be used for calculating the number of days between two date fields in a dataset.
| Benefit |
|---|
| 1. Easy to access accurate date-tools |
| Top Use Cases |
|---|
| 1. Calculate the duration a ticket was in a syterm |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform and make sure there are at least two date fields included in the dataset (e.g. date-raised and date-completed) 2. Run the Count number of days workflow |
Reformat Date Fields
Reformat Date Fields workflow helps you reformat date fields to the structure acceptable to Relevance AI (i.e. yyyy-mm-dd)
| Benefit |
|---|
| 1. Easy to access |
| Top Use Cases |
|---|
| 1. A preprocessing step for any date analysis such as time series |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run the Reformat Date Fields workflow when date values in your dataset do not follow yyyy-mm-dd |
Datasets
Connect to Appstore
Mine app reviews in AppStore using our scraping software and further analyze them with machine learning
| Benefits |
|---|
| 1. Easy access to recent and valuable data 2. Scrape customer feedback from any content in Appstore |
| Top Use Cases |
|---|
| Working with recent data from Appstore |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run the Connect to Appstore workflow |
Connect to Playstore
Mine app reviews in PlayStore using our scraping software and further analyze them with machine learning
| Benefits |
|---|
| 1. Easy access to recent and valuable data 2. Scrape customer feedback from any content in both Playstore |
| Top Use Cases |
|---|
| Working with recent data from Playstore |
| Start Analyzing |
|---|
| 1. Upload your dataset to Relevance AI's platform 2. Run the Connect to Playstore workflow |
Copy dataset
To create an exact copy of an existing dataset or copy a subset of a dataset.
| Benefit |
|---|
| Versioning and keeping track of changes |
| Top Use Cases |
|---|
| 1. Versioning 2. Sharing a subset of data |
| Start Analyzing |
|---|
| 1. Upload your data 2. Run Copy Dataset Workflow |
Migrate a Dataset from one project/region to another
| Benefit |
|---|
| Coming Soon |
| Top Use Cases |
|---|
| Coming Soon |
| Start Analyzing |
|---|
| Coming Soon |
Export Tags
Export Tags workflow allows you to format your Tags in an exported CVS file in a "One-Hot" structure. Under the one-hot structure, there will be one row per entry and one column per tag. Entries that are tagged with tag X will receive 1 under column X and 0 otherwise.
| Benefit |
|---|
| Export all or a subset of your data in addition to their tags under one-hot encoded format |
| Top Use Cases |
|---|
| Analysing data along with tags in softwares such as SPSS |
| Start Analyzing |
|---|
| 1. Upload your data 2.Run a tagging workflow such as AI-Tagging and Tag with your own tags 3. Export Tags workflow |
Export to xlsx
Export all or a subset of your dataset.
| Benefit |
|---|
| Export all or a subset of your data |
| Top Use Cases |
|---|
| Export a subset of data or the whole dataset to an Excel file |
| Start Analyzing |
|---|
| 1. Upload your data 2.Run Export Xlsx workflow |
Related Articles
Updated 5 months ago