Use cutting-edge NLU models from the comfort of Google Sheets. Write responses, select a model, select a ranking method, and send a query. Useful for bot making, games, and other semantic experiments.
|Apply for access||Private documentation|
- Businesses are using natural language understanding (NLU) to create digital personal assistants, customer service response bots, and semantic search engines for reviews, forums and the news.
- However, the perception that using NLU and machine learning is costly and time consuming prevents a lot of potential users from exploring its benefits.
- To dispel some of the intimidation of using NLU, and to demonstrate how it can be easily used with pre-trained, generic models, we have released a tool, called the Semantic Reactor.
Inputs and outputs:
User input steps:
- Create a corpus (list of candidate phrases) within a Google Sheet
- Select a model
- Select a ranking method
- Query the user-created corpus
Users receive back:
- Ranking of all the candidate phrases within the corpus
- Weights or scores of all the candidate phrases
Experiments to try:
- Test chatbot responses, for instance, customer service chatbots or marketing chatbots.
- Test word associations, for instance, if you wanted to make a game based on word associations.
- Test lists of headlines, titles, celebrity names, movie title or list of almost anything pertaining to any topic. Useful if you were interested in what topics or creative works are linked to specific topics or creators.
What data do I need?
Data and label types:
- The user just needs to create a corpus and start experimenting. No data other than that is needed.
- The corpus has to fit within a Google Sheet, so it can support thousands of lines, but it will slow down a bit the more you add rows.
What skills do I need?
- Familiarity with Google Sheets
- Familiarity with NLU concepts like semantic similarity