MongoDB Enables AI-Powered Legal Searches with Qura

Oliver Tree
September 18, 2024
#genAI

The launch of ChatGPT in November 2022 caught the world by surprise.

But while the rest of us marveled at the novelty of its human-like responses, the founders of Qura immediately saw another, more focused use case.

“Legal data is a mess,” said Kevin Kastberg, CTO for Qura. “The average lawyer spends tens of hours each month on manual research. We thought to ourselves, ‘what impact would this new LLM technology have on the way lawyers search for information?’”

And with that, Qura was born.

Gaining trust

From its base in Stockholm, Sweden, Qura set about building an AI-powered legal search engine.

The team trained custom models and did continual pre-training on millions of pages of publicly available legal texts, looking to bring the comprehensive power of LLMs to the complex and intricate language of the law.

“Legal searches have typically been done via keyword search,” said Kastberg. “ We wanted to bring the power of LLMs to this field. ChatGPT created hype around the ability of LLMs to write. Qura is one of the first startups to showcase their far more impressive ability to read. LLMs can read and analyze, on a logical and semantic level, millions of pages of textual data in seconds. This is a game changer for legal search.”

Unlike other AI-powered applications, Qura is not interested in generating summaries or “answers” to the questions posed by lawyers or researchers. Instead, Qura aims to provide customers with the best sources and information.

“We deliberately wanted to stay away from generative AI. Our customers can be sure that with Qura there is no risk of hallucinations or bad interpretation. Put another way, we will not put an answer in your mouth; rather, we give you the best possible information to create that answer yourselves,” said Kastberg.

“Our users are looking for hard-to-find sources, not a gen AI-summary of the basic sources,” he added.

With this mantra, the company claims to have reduced research times by 78% while surfacing double the number of relevant sources when compared to similar legal search products.

MongoDB in the mix

Qura has worked with MongoDB since the beginning.

“We needed a document database for flexibility. MongoDB was really convenient as we had a lot of unstructured data with many different characteristics.”

In addition to the flexibility to adapt to different data types, MongoDB also offered the Qura team lightning-fast search capabilities.

“MongoDB Atlas search is a crucial tool for our search algorithm agents to navigate our huge datasets. This is especially true of the speed at which we can do efficient text searches on huge corpuses of text, an important part for navigating documents,” said Kastberg.

And when it came to AI, a vector database to store and retrieve embeddings was also a real benefit.

“Having vector search built into Atlas was convenient and offered an efficient way to work with embeddings and vectorized data.”

What's next?

Qura's larger goal is to bring about the next generation of intelligent search. The legal space is only the start, and the company has larger ambitions to expand beyond Sweden and into other industries too.

“We are live with Qura in the legal space in Sweden and currently onboarding EU customers in the coming month. What we are building towards is a new way of navigating huge text databases, and that could be applied to any type of text data, in any industry,” said Kastberg.

Are you building AI apps? Join the MongoDB AI Innovators Program today! Successful participants gain access to free Atlas credits, technical enablement, and invaluable connections within the broader AI ecosystem. If your company is interested in being featured, we’d love to hear from you. Connect with us at ai_adopters@mongodb.com.

Head over to our quick-start guide to get started with Atlas Vector Search today.

← Previous

Top Use Cases for Text, Vector, and Hybrid Search

Search is how we discover new things. Whether you’re looking for a pair of new shoes, the latest medical advice, or insights into corporate data, search provides the means to unlock the truth. Search habits—and the accompanying end-user expectations—have evolved along with changes to the search experiences offered by consumer apps like Google and Amazon. The days of the standard of 10 blue links may well be behind us, as new paradigms like vector search and generative AI (gen AI) have upended long-held search norms. But are all forms of search created equal, or should we be seeking out the right “flavor” of search for specific jobs? In this blog post, we will define and dig into various flavors of search, including text, vector and AI-powered search, and hybrid search, and discuss when to use each, including sample use cases where one type of search might be superior to others. Information retrieval revolutionized with text search The concept of text search has been baked into user behavior from the early days of the web, with the rudimentary text box entry and 10 blue link results based on text relevance to the initial query. This behavior and associated business model has produced trillions in revenue and has become one of the fiercest battlegrounds across the internet . Text search allows users to quickly find specific information within a large set of data by entering keywords or phrases. When a query is entered, the text search engine scans through indexed documents to locate and retrieve the most relevant results based on the keywords. Text search is a good solution for queries requiring exact matches where the overarching meaning isn't as critical. Some of the most common uses include: Catalog and content search: Using the search bar to find specific products or content based on keywords from customer inquiries. For example, a customer searching for "size 10 men trainers" or “installation guide” can instantly find the exact items they’re looking for, like how Nextar tapped into Atlas Search to enable physical retailers to create online catalogs Covid-19 pandemic. In-application search: This is well-suited for organizations with straightforward offerings to make it easier for users to locate key resources, but that don’t require advanced features like semantic retrieval or contextual re-ranking. For instance, if a user searches for "songs key of G," they can quickly receive relevant materials. This streamlines asset retrieval, allowing users to focus on the task they are trying to achieve and boosts overall satisfaction. For a company like Yousician , Atlas Search enabled their 20 million monthly active users to tackle their music lessons with ease. Customer 360: Unifying data from different sources to create a single, holistic view. Consolidated information such as user preferences, purchase history, and interaction data can be used to enhance business visibility and simplify the management, retrieval, and aggregation of user data. Consider a support agent searching for all information related to customer “John Doe." They can quickly access relevant attributes and interaction history, ensuring more accurate and efficient service. Helvetia was able to achieve success after migrating to MongoDB and using Atlas Search to deliver a single, 360-degree real-time view across all customer touchpoints and insurance products. AI and a new paradigm with vector search With advances in technology, vector search has emerged to help solve the challenge of providing relevant results even when the user may not know what they’re looking for. Vector search allows you to take any type of media or content, convert it into a vector using machine learning algorithms, and then search to find results similar to the target term. The similarity aspect is determined by converting your data into numerical high-dimensional vectors, and then calculating the distance between them to determine relevance—the closer the vector, the higher the relevance. There is a wide range of practical, powerful use cases powered by vector search—notably semantic search and retrieval-augmented generation (RAG) for gen AI. Semantic search focuses on meaning and prioritizes user intent by deciphering not just what users type but why they're searching, in order to provide more accurate and context-oriented search results. Some examples of semantic search include: Content/knowledge base search: Vast amounts of organizational data, structured and unstructured, with hidden insights, can benefit significantly from semantic search. Questions like “What’s our remote work policy?” can return accurate results even when the source materials do not contain the “remote” keyword, but rather have “return to office” or “hybrid” or other keywords. A real-world example of content search is the National Film and Sound Archive of Australia , which uses Atlas Vector Search to power semantic search across petabytes of text, audio, and visual content in its collections. Recommendation engines: Understanding users’ interests and intent is a strong competitive advantage–like how Netflix provides a personalized selection of shows and movies based on your watch history, or how Amazon recommends products based on your purchase history. This is particularly powerful in e-commerce, media & entertainment, financial services, and product/service-oriented industries where the customer experience tightly influences the bottom line. A success story is Delivery Hero , which leverages vector search-powered real-time recommendations to increase customer satisfaction and revenue. Anomaly detection: Identifying and preventing fraud, security breaches, and other system anomalies is paramount for all organizations. By grouping similar vectors and using vector search to identify outliers, potential threats can be detected early, enabling timely responses. Companies like VISO TRUST and Extrac are among the innovators that build their core offerings using semantic search for security and risk management. With the rise of large language models (LLMs), vector search is increasingly becoming essential in gen AI application development. It augments LLMs by providing domain-specific context outside of what the LLMs “know,” ensuring relevance and accuracy of the gen AI output. In this case, the semantic search outputs are used to enhance RAG. By providing relevant information from a vector database, vector search helps the RAG model generate responses that are more contextually relevant. By grounding the generated text in factual information, vector search helps reduce hallucinations and improve the accuracy of the response. Some common RAG applications are for chatbots and virtual assistants, which provide users with relevant responses and carry out tasks based on the user query, delivering enhanced user experience. Two real-world examples of such chatbot implementations are from our customers Okta and Kovai . Another popular application is using RAG to help generate content like articles, blog posts, scripts, code, and more, based on user prompts or data. This significantly accelerates content production, allowing organizations including Novo Nordisk and Scalestack to save time and produce content at scale, all at an accuracy level that was not possible without RAG. Beyond RAG, an emerging vector search usage is in agentic systems . Such a system is an architecture encompassing one or more AI agents with autonomous decision-making capabilities, able to access and use various system components and resources to achieve defined objectives while adapting to environmental feedback. Vector search enables efficient and semantically meaningful information retrieval in these systems, facilitating relevant context for LLMs, optimized tool selection, semantic understanding, and improved relevance ranking. Hybrid search: The best of both worlds Hybrid search combines the strengths of text search with the advanced capabilities of vector search to deliver more accurate and relevant search results. Hybrid search shines in scenarios where there's a need for both precision (where text search excels) and recall (where vector search excels), and where user queries can vary from simple to complex, including both keyword and natural language queries. Hybrid search delivers a more comprehensive, flexible information retrieval process, helping RAG models access a wider range of relevant information. For example, in a customer support context, hybrid search can ensure that the RAG model retrieves not only documents containing exact keywords but also semantically similar content, resulting in more informative and helpful responses. Hybrid search can also help reduce information overload by prioritizing the most relevant results. This allows RAG models to focus on processing and understanding the most critical information, leading to faster, more accurate responses, and improving the user experience. Powering your AI and search applications with MongoDB As your organization continues to innovate in the rapidly evolving technology ecosystem, building robust AI and search applications supporting customer, employee, and stakeholder experiences can deliver powerful competitive advantages. With MongoDB, you can efficiently deploy full-text search , vector search , and hybrid search capabilities. Start building today—simplify your developer experience while increasing impact in MongoDB’s fully-managed, secure vector database, integrated with a vast AI partner ecosystem , including all major cloud providers, generative AI model providers, and system integrators.

September 16, 2024

Next →

Ahamove Rides Vietnam’s E-commerce Boom with AI on MongoDB

The energy in Vietnam’s cities is frenetic as millions of people navigate the busy streets with determination and purpose. Much of this traffic is driven by e-commerce, with food and parcel deliveries perched on the back of the country’s countless motorcycles or in cars and trucks. In the first quarter of 2024, online spending in Vietnam grew a staggering 79% over the previous year. Explosive growth like this is expected to continue, raising the industry’s value to $32 billion by 2025 , with 70% of the country’s 100 million population making e-commerce transactions . With massive numbers like this, in logistics, efficiency is king. The high customer expectations for rapid deliveries drive companies like Ahamove to innovate their way to seamless operations with cloud technology. Ahamove is Vietnam’s largest on-demand delivery company, handling more than 200,000 e-commerce, food, and warehouse deliveries daily, with 100,000 drivers and riders plying the streets nationwide. The logistics leader serves a network of more than 300,000 merchants, including regional e-commerce giants like Lazada and Shopee, as well as nationwide supermarket chains and small restaurants. The stakes are high for all involved, so maximizing efficiency is of utmost importance. Innovating to make scale count Online shoppers’ behavior is rarely predictable, and to cope with sudden spikes in daily delivery demand, Ahamove needed to efficiently scale up its operations to enhance customer and end-user satisfaction. Moving to MongoDB Atlas on Amazon Web Services (AWS) in 2019, Ahamove fundamentally changed its ability to meet the rising demand for deliveries and new services that please e-commerce providers, online shoppers, and diners. The scalability of MongoDB is crucial for Ahamove, especially during peak times, like Christmas or Lunar New Year, when the volume of orders surges to more than 200,000 a day. “MongoDB's ability to scale ensures that the database can handle increased loads, including data requests, without compromising performance and leading to quicker order processing and improved user experience,” said Tien Ta, Strategic Planning Manager at Ahamove. One of the powerful services that improves e-commerce across Vietnam is geospatial queries enabled by MongoDB. Using this geospatial data associated with specific locations on Earth's surface, Ahamove can easily locate drivers, map drivers to restaurants to accelerate deliveries, and track orders without relying on third-party services to provide information, which slows deliveries. Meanwhile, the versatility of MongoDB’s developer data platform empowers Ahamove to store its operational data, metadata, and vector embeddings on MongoDB Atlas and seamlessly use Atlas Vector Search to index, retrieve, and build performant generative artificial intelligence (AI) applications. AI evolution Powered by MongoDB Atlas , Ahamove is transforming Vietnam’s e-commerce industry with innovations like instant order matching, real-time GPS vehicle tracking, generative AI chatbots, and services like driver rating and variable delivery times, all available 24 hours a day, seven days a week. In addition to traffic, Vietnam is also famous for its excellent street food. Recognizing the importance of the country’s rapidly growing food and beverage (F&B) industry, which is projected to be worth more than US$27.3 billion in 2024 , Ahamove decided to help Vietnam’s small food vendors benefit from the e-commerce boom gripping the country. Using the latest models, including ChatGPT-4o-mini and Llama 3.1, Ahamove’s fully automated generative AI chatbot on MongoDB integrates with restaurants’ Facebook pages. This makes it easier for hungry consumers to handle the entire order process with the restaurant in natural language, from seeking recommendations to placing orders, making payments, and tracking deliveries to their doorsteps. How AhaFood AI chatbot automate food order journey “Vietnam’s e-commerce industry is growing rapidly as more people turn to their mobile devices to purchase goods and services,” added Ta. “With MongoDB, we meet this customer need for new purchase experiences with innovative services like generative AI chatbots and faster delivery times.” Anticipated to achieve 10% of food deliveries at Da Nang market and take the solution nationwide in the first half of 2025, AhaFood.AI - Ahamove’s latest initiative, also provides personalized dish recommendations based on consumer demographics, budgets, or historical preferences, helping people find and order their favorite food faster. Moreover, merchants receive timely notifications of incoming orders via the AhaMerchant web portal, allowing them to start preparing dishes earlier. AhaFood.AI also collects and securely stores users’ delivery addresses and phone numbers, ensuring better driver assignment and fulfilling food orders in less than 15 minutes. “Adopting MongoDB Atlas was one of the best decisions we’ve ever made for Ahamove, allowing us to build an effective infrastructure that can scale with growing demand and deliver a better experience for our drivers and customers,” said Ngon Pham, CEO, Ahamove. “Generative AI will significantly disrupt the e-commerce and food industry, and with MongoDB Vector Search we can rapidly build new solutions using the latest database and AI technology.” The vibrant atmosphere of Vietnam's bustling cities is part of the country's charm. Rather than seeking to bring calm to this energy, Vietnam thrives on it. Focusing on improving efficiency and supporting street food vendors in lively urban areas with cloud technology will benefit all. Learn how to build AI applications with MongoDB Atlas . Head over to our quick-start guide to get started with Atlas Vector Search today.

September 19, 2024