Enhance Qlik Sense with DeepSeek AI: A Game of Thrones Example

DeepSeek represents a significant advancement in Large Language Models (LLMs), offering improved accuracy and applicability across various contexts. Integrating DeepSeek with Qlik Sense allows organizations to enhance their analytical capabilities and decision-making processes by leveraging AI-driven insights. This article will guide you through the process of integrating DeepSeek's APIs with Qlik Sense, using a practical example to illustrate the benefits.

DeepSeek's APIs provide access to a range of AI functionalities, including natural language processing (NLP), computer vision, and data analysis. These APIs enable developers and businesses to integrate advanced AI capabilities into their existing systems and applications. The DeepSeek-R1 model, a Large Reasoning Model (LRM), excels in logical reasoning and outperforms other advanced LLMs in math and reasoning benchmarks. Accessing the DeepSeek API requires an API key, which can be obtained from the DeepSeek platform. Note that API usage may incur costs based on the volume of calls or processing.

To demonstrate the integration of Qlik Sense with DeepSeek, we will use a Game of Thrones (GOT) application. This application displays information about the main characters from the series, sourced from a public API (https://thronesapi.com/api/v2/Characters). While the API provides basic information such as name, title, and family, it lacks detailed character descriptions. We will leverage DeepSeek to generate these descriptions and enrich the character data within Qlik Sense.

First, we need to load the GOT character data from the public API. This can be achieved using the REST connector in Qlik Sense. Create a new REST connection named 'got' using the GET method, as the API does not require authentication. Then, create a script to read the API and load the character data into a Qlik Sense table. This will provide a list of characters with their names, titles, families, and photos.

To obtain character descriptions, we will integrate with the DeepSeek API using the 'deepseek-chat' method. The DeepSeek API is similar to OpenAI's API, allowing you to use OpenAI libraries in languages like Python and NodeJS. Create a REST connection to DeepSeek using the POST method. Configure the headers with 'Content-Type: application/json' and 'Authorization: Bearer <Your API-Key>'. Adjust the temperature parameter to control the creativity of the generated descriptions. Ensure you have sufficient credit in your DeepSeek account to avoid API errors.

To generate a description for each character, we will use the following prompt: 'Make a brief description about the GOT character <character name>'. Loop through the table containing the character data (GOTCharascters) and create the prompt text with the character's name (FullName field). Then, request the DeepSeek API through the REST connector to generate the description. Load the generated description into a new table, linking it to the character ID to maintain the relationship between the character data and the AI-generated descriptions. This will result in two tables: one with the list of characters and another with the summarized descriptions.

Qlik Data Flow, integrated into Qlik Cloud Analytics, simplifies and accelerates data preparation for analytics and AI applications. Using its intuitive drag-and-drop interface, you can combine, clean, and transform datasets without coding. You can use Data Flow to capture character data and prepare it for integration with DeepSeek. This NoCode approach makes it accessible to users regardless of their technical level.

Integrating DeepSeek's API with Qlik Sense enhances data analysis and utilization, bringing automation, accuracy, and deeper insights. This combination allows you to go beyond traditional analytics, exploring the potential of AI to drive innovation and strategic decision-making. By enriching data with AI-generated content, you can gain a more comprehensive understanding of your data and make more informed decisions.