LinkedIn

Twitter

GitHub

Learn how to deliver conversational analytics using large language models, and how to integrate our GenAI Chatbot components for user-friendly analytical insights.

genai-hero-img-1-min

GenAI-Hero-Img (1)-min

Conversational analytics with generative AI

Conversational analytics

Analytics Chatbot

GenAI chat components

speakers-3-1

Speakers (3) (1)

GenAI, in-context analytics, and the future of data monetization

Generative AI, or GenAI, refers to a type of artificial intelligence that is capable of creating new content, whether that be text, images, music, or other forms of digital media. It works by learning from a large dataset of existing content and then using that knowledge to generate new, original material that is often indistinguishable from human-created content. <a href="https://openai.com/blog/chatgpt">ChatGPT</a>, launched publicly by OpenAI in late 2022, is a popular example of generative AI in a conversational ‘chat’ user interface. By leveraging vast training data and powerful natural language processing capabilities, GenAI can help businesses and individuals interact with data more efficiently, solve problems, and access information. Leading embedded analytics companies, including Sisense, are using GenAI to augment analytics, generate data storytelling narratives, recommend areas to explore, suggest samples, and build conversational analytics functionality like assistants and chatbots.

What is generative AI?

Sisense’s Analytics Chatbot is an embeddable analytics conversational chat tool that allows business users to explore data for insights through natural language questions powered by GenAI. The Analytics Chatbot can suggest sample insights, provide narrative explanations in context, and produce data visualizations all through a common chat interface. The chatbot is available as a component in <a href="https://sisense.dev/guides/sdk/">Compose SDK</a>. It is assembled from functional building blocks that are also available as part of Compose SDK: an analytics chat interface, quick-start insights, and data storytelling.

What is Sisense Analytics Chatbot?

You can create highly customized GenAI analytics interfaces by using any of the individual components that make up the Analytics Chatbot in the combination of your choice to provide the functionality you want. Our analytics chat interface, quick-start insights, and data storytelling are components available in <a href="https://sisense.dev/guides/sdk/">Compose SDK</a> along with the pre-made Analytics Chatbot.

What other ways can I create a conversational AI tool for my product? 

<a href="https://docs.sisense.com/main/SisenseLinux/simply-ask-query-in-natural-language.htm">Simply Ask</a> provides self-service data exploration via a Natural Language Querying interface. It supports semantic similarity and synonyms by leveraging a sentence transformer language model. It is available through our Fusion Embed application. 
The GenAI Analytics Chatbot offers much more. It has a conversational interface where you can ask questions and follow-up questions. It provides AI-generated in-context, quick-start questions to guide the conversation, and narratives of the data for additional context. The Analytics Chatbot is powered by a current generation of LLM. It is packaged as part of <a href="https://sisense.dev/guides/sdk/">Compose SDK</a> and provides a higher level of customization.

How is the Analytics Chatbot different from the Simply Ask feature in Sisense Fusion?

The Sisense platform complies with SOC 2 and ISO 27001 frameworks and is HIPAA-ready. Sisense has also obtained an ISO27701 (for PII Processors) certification, which serves as a third-party assurance that Sisense processes data in accordance with some of the strict data privacy regulations (such as the GDPR and CCPA). Customers who access the GenAI services via Sisense’s OpenAI Azure license key can rest assured that our data privacy and security standards continue to apply. 
For more information please see Sisense’s Privacy and Data Protection Center and <a href="https://www.sisense.com/platform/security-and-trust/">Security and Trust Center</a>.

What data privacy and security controls are in place?

Not when using the Analytics Chatbot with the OpenAI Azure license key provided by Sisense. Customers who use the Analytics Chatbot using their their own GenAI license should refer to the terms and conditions of that license.

Will chat interactions become LLM training data?

Sisense Analytics Chatbot works directly with the same Sisense <a href="https://docs.sisense.com/main/SisenseLinux/choosing-the-right-data-model.htm">data models</a> and semantic layer that also drive all your Sisense dashboards, reporting, and analysis. This ensures your users get the same consistent fields, metrics, and relationships for their conversational analytics chat experience, as their other analytics experiences. This consistency ensures both trust, as well as providing a single foundation, with the same row and column level security across all experiences.

Can Sisense Analytics Chatbot handle my data schema?

The results of all Sisense Analytics Chatbot queries are driven by the Sisense query engine and a robust, trusted data model. It enables analytics delivery teams to centrally define the scope, meaning, fields, and relationships that encompass the end-user analytics experience. Users can easily select the data topic and the data model that they want to investigate through a conversational analytics experience directly in the Analytics Chatbot.

How do I know if the answers generated by the Analytics Chatbot are accurate?

Sisense’s semantic layer makes it easy to get your data “AI-ready”. Start by reviewing your columns and metrics in the model to ensure they are easy to understand, clearly defined, and natural for your users to have a conversational analytics experience with. With Perspectives, you can easily create a controlled view of your full data model, and control the context for business users to self-serve analytics.

What can I do to improve the quality of the results?

Frequently asked questions

Explore other resources

Data analytics powered by generative AI

conversational-analytics-2

Conversational-Analytics 2

The top analytics use cases for GenAI are to recommend places to begin data exploration, augment analytics, conversationally query data, and achieve narrative answers. Sisense’s Analytics Chatbot (beta release, available in private preview) gives you quick and intuitive access to the power of GenAI for analytics. And because it’s composable, you can customize the perfect chat experience for your users.

Unlock the power of conversational analytics

analytics-chatbot-min

Analytics-Chatbot-min

Want to help your users ask the right questions? Empower them to <a href="https://sisense.com/blog/add-gen-ai-powered-analytics-to-apps/">explore the data and self-serve insights</a>, all using natural language. Then add Compose SDK’s Analytics Chatbot to your product. The chatbot allows users to explore data through a conversational chat interface and provides avenues of exploration and narrative explanations of charts and graphs to gain a quick understanding of the insights. The Analytics Chatbot is currently in private preview with an upcoming beta release.

Integrate it your way with Compose SDK

genai-chat-components-min

GenAI-Chat-Components-min

Build custom GenAI experiences with <a href="https://sisense.dev/guides/sdk/">Compose SDK</a>. With flexible React Components and APIs, mix and match our GenAI building blocks to create your own tailored, LLM-powered Analytics Chatbot or use them as discrete components within your products.
<ul>
<li>Deliver in-app analytics chat 
Business users simply ask questions in a conversational interface to uncover data insights. </li>
<li>Provide easy quick-start insights  
The right questions uncover actionable answers. GenAI suggests starting points to explore the data landscape.</li>
<li>Bring insights to life with data storytelling 
Auto-generate narratives to add context and enhance data storytelling and collaboration.</li>
</ul>