Once the documents are accurately classified, data is then extracted and the system will automatically route them to the appropriate systems and/or staff within the organisation.
Automatically transform and extract data from a wide range of documents, including hard-to-process unstructured documents, which enables more effective and insightful decisions across a larger number of critical high-value business processes.
Build and deploy more document transformation projects quicker and with less upfront investment to maximise the return on your investment.
Capture more documents and information with the same resources.
Capture documents and information much quicker, thereby accelerating business processes for faster response times.
Reduce manual labour and total headcount required for manual document classification, separation and data entry.
Increase information quality and reduce exceptions in information-intensive business processes.
Providing enhanced support for hybrid digital mailroom solutions by classifying and extracting PDF documents created from faxes and office documents using Tungsten Capture import connectors. The system utilises the PDF text layer so that the need for OCR is eliminated. With this capability, organisations can deploy Tungsten Transformation Modules as a single solution to handle all documents, paper or electronic.
This allows different configurations for different processing steps, such as classification and extraction of forms, and extraction of correspondence. This thereby reduces solution complexity, configuration and maintenance costs. Each configuration can be created and maintained separately so the solution is easier to manage.
Supports parallel development of configuration files, which allows multiple developers to work on the same solution at the same time. This greatly reduces timescales and allows specialised skills to be leveraged more efficiently.
Provides a new separation benchmarking tool and improves the classification and extraction benchmarking tools by adding the ability to easily compare results before and after making changes to classification settings. The benchmarking data serves administrators in determining if performance has improved – This capability helps to ensure that the highest accuracy is achieved and maintained on all document types, thereby increasing ROI.
A validation client allows users to attach “sticky notes” to documents that pass information and comments to users of downstream business applications, therefore improving efficiency by reducing the need for emails, phone calls or other communications.
This supports multi-purchase order invoices that will increase automatic matching and straight-through processing rates, thereby reducing processing costs and providing additional learning options.
This supports matching and validation of extracted data against databases containing millions of records using a dedicated server. This also features super-fast query response support for very large databases with 64-bit support, automatic update of data sources, load balancing across multiple servers, automatic scheduled database updates, and other high availability features.
This supports custom layouts and buttons, domain login, single sign-on, and annotations which preserves user defined settings such as username at login screen and UI settings between sessions and provides a comprehensive set of batch editing options; all to improve ease of use, user experience, productivity and processing speed.
Provides a new tool to identify and resolve conflicts between documents used for extraction training, thereby improving learning and performance and delivering greater extraction accuracy without the need to add any new training samples.
Clustering capability for Tungsten Transformation 6.0 minimises the time to locate, identify, organise, label and file samples of each document type used for Tungsten Transformation training. This mean that if you point it at a large directory of unsorted, unstructured document images, after a few mouse clicks it produces an organised file/folder training set you can use for content classification learning in Tungsten Transformation Project Builder. What used to take days or weeks will take only minutes!
This provides a new docking feature that can place the document image at the left, right, top or bottom to make maximum use of screen real estate and user workflow patterns. This also enables users to override the default image position and zoom settings to suit their preference and preserve their settings for next time.
Provides new project settings to make it easier to provide a fully localised interface for users; including both menu items and labels.
By reducing and, in some applications, removing the need for manual document classification, separation and data extraction, organisations can drastically reduce their operational costs – The higher the level of Touchless Processing, the greater the cost reduction. With Tungsten Transformation, users will achieve high levels of automation on any document type.
Manual document classification, document separation and data extraction is time-consuming, tedious, and a poor use of valued staffs’ time. By automating these processes, Tungsten Transformation allows knowledge workers to focus their efforts on value-adding tasks such as customer service or query resolution. Speed of processing is also increased when processes become touchless.
By removing the need for manual document classification, separation and extraction, companies are able to process more transactions each day whilst maintaining the same staffing levels. With Tungsten Transformation, users achieve high levels of productivity on any document type.
Manual document classification, separation and data extraction is error-prone; humans typically achieve a peak accuracy rate of 95%. By automating these processes and tuning the technologies on a client documents, Tungsten Transformation can improve the quality of extracted data compared to manual keying (typically reducing errors to less than 5%) whilst maintaining a high degree of automation. Also, the powerful data validation capabilities of Tungsten Transformation ensure that extracted information is correct before it is used by business systems to assist in decision making processes. The result is fewer processing exceptions compared to manual methods, fewer customer queries and more accurate and timely decisions.
Tungsten Transformation provides a controlled IT-based framework for gathering sensitive customer data as soon as it enters an organisation and delivers it quickly into a secure repository. Powerful data validation capabilities ensure that the data is correct. Accurate document classification means that content retention and security policies can be applied more reliably.
Tungsten Transformation can classify, separate and extract a document in a matter of seconds; also running outside normal business hours if necessary. Consequently, documents can be processed much quicker than manual methods therefore reducing the time to complete each transaction and benefiting both the user and customer.
Automatic document processing reduces the need for human interaction with confidential data hence increasing security. By providing the capability to automatically classify documents as they are scanned and feed them into a workflow system for automatic routing, Tungsten Transformation removes the need for mailroom operators to read documents at the point of receipt in order to send them to the correct department. Consequently, the risk of confidential information being exposed in the public domain is decreased. After extraction, sensitive data can be identified and routed only to key personnel, allowing customer privacy and confidentiality issues to be addressed, resulting in increased customer confidence and better adherence to regulatory compliance.
Manual processes are intrinsically difficult to audit. Using automated technology to process documents means that every step can be controlled, monitored and reported on. As a result, organisations can better meet the requirements of regulatory compliance and bottlenecks can be identified more easily.
Tungsten Transformation is the most complete and versatile document classification and data extraction offered on the market. It can process structured, semi-structured and unstructured documents – including hand-printed and handwritten forms, invoices, checks, correspondence and any other document type, all on the same platform. Legacy documents can also be processed, thus avoiding expensive and time consuming form redesign processes and allowing efficient back file conversion.
Tungsten Transformation features powerful learn-by-example technologies for classification, separation and extraction that can be configured quickly and cheaply compared to traditional rules based methods. Moreover, as documents are processed in a production environment, on-line learning can be used to ‘teach’ the system characteristics of problem documents so they are processed automatically in the future, increasing ROI and reducing TCO since no IT skills are needed.
In an advanced recognition environment, easy to use and efficient data validation is vital for maintaining productivity: Poor validation design can cancel out the benefits of automated classification and extraction. Tungsten Transformation’s keying applications are highly customisable, either at the design time through an easy to use drag and drop interface or during production operations through a comprehensive set of script events. During production, screen navigation and image manipulation are logical and easy to use. Separate interfaces are provided for reviewing document classification, single character correction and data validation, and at each stage the interface can be tuned for maximum efficiency.
Tungsten Transformation’s user interfaces are highly customisable to meet customer specific requirements in terms of screen layout and interface behaviour. They can also be configured to mimic legacy systems which will reduce training costs and speed up the adoption of new systems.
As documents are processed Tungsten Transformation gathers detailed information regarding recognition rates and presents statistics per field, document types, and average over user selectable time periods. This allows users to monitor system performance and also get early notification of any and all issues.
Tungsten Transformation integrate seamlessly with Tungsten Capture, giving access to the widest range of document scanners and back-end storage solutions. Tungsten Transformation also benefits from Tungsten Capture’s distributed nature and add-on products including Tungsten Front office Server, Tungsten Web Capture and Tungsten Mobile Capture. Tungsten Capture also provides high availability and enterprise capabilities to easily scale transformation capabilities horizontally and vertically throughout any size organisation. Powerful advanced recognition capabilities can also be added to any Tungsten Capture installation without any disruption to your existing infrastructure.