ReportMiner 11.1 - Release Notes
Last updated
Was this helpful?
Last updated
Was this helpful?
ReportMiner 11.1 marks our step into the realm of AI, empowering users to harness the full potential of data extraction and processing with newly added AI capabilities. From user interface improvements to advanced AI-powered features, this release sets a new standard for template-less data extraction. Template-less data extraction uses AI and machine learning to pull data from documents without the need to use pre-configured templates, making it easier to handle different and unorganized document layouts.
ReportMiner 11.1 sets a new benchmark in efficiency and usability, seamlessly integrating AI into your data workflows.
Elevate your data journey with ReportMiner 11.1 – where performance, visibility, and efficiency converge effortlessly, all within an intuitive drag-and-drop interface.
LLM Generate is a core component of Astera’s AI capabilities, enabling the creation of AI-powered solutions when combined with other objects on the flow, including sources, transformations, and destinations. It retrieves an output from a Large Language Model based on a user-defined input prompt, with support for various LLM providers such as OpenAI, Llama, and custom models.
The object features:
Input Port: Maps fields to be included in the prompt for the LLM model.
Output Port: Contains the generated result from the LLM model.
LLM Generate’s flexibility in processing input and generating output through natural language instructions makes it a versatile and powerful tool in your data pipeline. Numerous use cases are made possible thanks to this new object, which will be explored in the product documentation.
The Text Converter object is another addition to the dataflow’s toolbox, which enables users to extract text from various file formats, including documents, images, and scanned files. It enhances text extraction performance using Optical Character Recognition (OCR) technology. Currently, the Text Processor supports Google OCR, PaddleOCR (Beta), TesseractOCR (Beta), and TextractOCR (Beta) platforms.
Key conversion features include:
Document to Text: Extract text from PDFs, Doc/DocX, and TXT files.
Image to Text: Use OCR to extract text from image formats such as JPG, PNG, and JPEG.
HTML to Text: Extract text from HTML, HTM, and XHTML files.
Markdown to Text: Extract text from MD, MARKDOWN, MKD, MKDN, MDWN, and MDOWN files.
Excel to Text: Extract text from XLS, XLSX, and CSV files.
Astera ReportMiner's Auto-Generate Layout (AGL) feature uses AI to automatically identify data regions and fields in your source document, making it easier to create layouts in different document types and extract data.
Normally, creating a template can take over 10 minutes, but with the AGL feature, it could take as little as 5 seconds. This feature also checks the extracted data for accuracy identifying any errors or issues that may require user’s review.
This helps speed up data extraction with less manual work involved, allowing the user to focus on other tasks.
To enable advanced extraction features, such as the Text Converter, the Python server is required. As part of the installation, the Python server is embedded within the integration server in v11.1, and it runs seamlessly as a component within the server. Python server activation is required to utilize the Text Converter object within the tool.
The Install Manager now includes the Python Server installation as well for features like Text Converter and more. The Python Server must be installed on the same machine as the Integration Server, and the Install Manager will launch automatically if selected during the installation. If the user doesn't open the install manager upon server installation (via Wix installer), they can launch it from the Start menu later on.
With the WiX installer, users can customize their installation by choosing the installation directory and modifying the service port for the server installer, offering greater flexibility during setup compared to the previous versions. The v11.1 installation is designed to run alongside any pre-11 versions, so you can test your existing flows on either version side-by-side while transitioning to the new version. While we recommend setting up a new repository database for your new 11.1 installation, upgrading an existing repository is also supported.
This concludes ReportMiner 11.1 release notes.
