Curated User Experiences and Search Modalities

Learn How Curated User Experiences and Search Modalities accelerate User Adoption of your Enterprise Search Solution

Asset 1

Introduction

At Aretec, we've learned firsthand the importance of advanced enterprise search solutions, particularly in regulated environments where information accuracy, security, and compliance are paramount. Through our experience working with Enterprise Search powered by Large Language Models (LLMs) tools revolutionize the way employees access and utilize information, empowering them to work smarter, collaborate more effectively, and make informed decisions with confidence. By leveraging capabilities such as semantic search, LLM-driven summaries, intent detection, and boosting, while adhering to strict regulatory requirements, modern enterprise search solutions are transforming the search experience and driving productivity to new heights. 

In this blog, we will explore the key features and benefits of different modalities so users can receive the critical information they require and make data driven decisions. Some cases all the search capabilities can be embedded into websites, chat tools such as MS teams and Google chat, and existing applications.

Search Modalities

Before we review the different modalities, I want to explain some of the main features we talk about in this post. 

  • 1. Semantic Search

    • Semantic search understands query context and intent, improving result relevance.
    • Benefit: Reduces search time, increases productivity, and fosters knowledge discovery.

  • 2. LLM Driven Summary

    • LLM-driven summaries provide concise overviews of search results.
    • Benefit: Saves time by allowing quick relevance assessment of lengthy documents.

  • 3. Citations

    •  Citations provide references to original information sources.
    • Benefit: Maintains information integrity and trust, ensures reliable data for decision-making.

  • 4. Intent Detection

    •  Intent detection understands the purpose behind a search query.
    • Benefit: Provides targeted results, improves search experience, and reduces search time.

  • 5. Manual Results Sorting/Filtering

    •  Manual sorting and filtering allow customization of search results based on specific criteria.
    • Benefit: Gives employees control to narrow down information quickly, especially for large datasets.

  • 6. Follow-up Questions

    •  Follow-up questions enable iterative searching without starting from scratch
    • Benefit: Refines search results, finds specific information, and enhances problem-solving.

  • 7. Boosting (relevance/date)

    •  Boosting adjusts relevance scores based on factors like query relevance and publication date.
    • Benefit: Helps find pertinent and current information quickly, critical for informed decision-making.

  • 8. Search Results

    • Search Results are returned as part of the search and pagination is implemented to allow users to navigate all search results. Generally these are shown as blue links.
    • Benefit: Allows users to navigate thru all search results regardless of the number of results.

Below is a table that lays out the features and lists capabilities available in each modality

Table1

Now lets get into the different search modalities.

Traditional Search Experience

Goals: Provide a search interface with a simple search bar and concise results

User Story: As an Enterprise Search User, I need to get the most relevant and accurate results so that I can review the results and make my decisions.

Personas: General Users, Website Search Users, General Internal Search.

Usage: Website Search, General Knowledge Base Search,

Cost: $

Users are provided information in the form of sliders or lists and relevant information is brought to the top

Search bar is added to existing webpages and a list of results is returned based on the search

Example Link:

https://www.worldbank.org/en/search?q=health%20project%20details&Type=All

 This search experience uses Google 

Worldsearch

Traditional Search Experience with Summary

Goals: Provide an LLM generated search summary in addition to list of search results. Ability to ask follow up questions and maintain context some advanced search

User Story: As an Enterprise Search User, I need a relevant summary of the top results along with blue links so that I can review the summary along with the blue links.

Personas: General Users, Website Search Users, Researchers, General Internal Search.

Usage: Website Search, General Knowledge Search,

Cost: $$

Traditionally, the internal collections are not blended with external information to provide information or answers based on Public Data.

Embedded Search Chatbot

Goals: Provide a search experience within a team collaboration tools such as google chat, slack or MS Teams or websites. Allows users to get to search results and get very concise results with citations.

User Story: As an Enterprise Search User, I need a relevant summary with citations in a chatbot experience so that I can get decision information fast.

Personas: General Internal Search, chatbot users.

Usage: Helpdesk Search, Policy Document Search, Controlled Search, General Knowledge Base Search,

Cost: $$

Summary: Users are provided a short summary and top citation within the tools.

Chatnewone

Hybrid Search Experience

Goals: Developed by the Aretec team to show a unified user experience, combining Search and ChatGPT like summarization capabilities. 

User Story: As an Enterprise Search User, I need a relevant summary with citations broken down by in a chatbot experience so that I can get decision information fast.

Personas: General Internal Search, chatbot users.

Usage: Website Search, General Knowledge Base Search, Blended Search, 

Cost: $$$

Summary: Has access to Vertex AI Search and summarizes results while also providing citations from different data collections 

Sear 1

Gemini or Chatgpt Like Search

Goals: Search thru the corpus of documents and return detailed summaries. The focus is to provide very detailed summaries with some citation links. 

User Story: As a search user, I want detailed summaries of semantic searches so that I can use the content for decision making.

Personas: General Users, Researchers, General Internal Search.

Usage: Website Search, General Knowledge Base Search, 

Cost: $$

 Summary: Has access to Vertex AI Search and summarizes results while also Limited citations 

Seacfd

Submission Successful

The form has been successfully submitted.

Cookie 🍪
We use cookies to improve your experience on our site.
//chat// button.fixed{ display:none;}