 |
LanguageWare is the new generation IBM linguistic platform. It was designed from the ground up to address the demands posed by today's global applications. |
|
|
| Our light-weight component technology can be easily embedded into any existing solution to provide natural language understanding transparently and unobtrusively for the end user. LanguageWare is already embedded in a wide variety of applications - from e-mail to instant messaging, databases to business intelligence solutions and across many platforms from the desktop to the mainframe.
Although the concrete business problems that can be solved by the integration of such NLP technology are endless, here are some examples of where LanguageWare can make a true difference to how you run your business. |
|
|
| For most hospitals, the knowledge that resides within their unstructured text (referral or discharge letters, clinical notes, etc.) is largely ignored by their operational processes. The challenges are: to apply text analysis to convert this largely inaccessible textual information into structured data, which could then be linked with other data sources across the hospital system; to use this new enriched information to drive healthcare business processes, such as patient risk assessment; and finally identify how the results could be integrated into the more traditional operational processes, such as emergency room admission. The solution requires a holistic approach to the problem definition working side-by-side with the hospital information technology, healthcare, and admission staff, and in proposing new operational standards and processes that ultimately demonstrate the value such an integrated text-driven strategy can realize. |
|
|
| Considering that much of the work of investigating a crime is talking with people, asking questions, getting answers, and making hypotheses based on this information, it isn't surprising that there is a lot of text moving through various police departments and intelligence agencies around the world. The challenge in this project is to analyze these large volumes of unstructured text, extract references to things (people, places, dates) and relationships between them (person and their date-of-birth), build comprehensive triple stores that allow deep analysis and inferencing, and connect this data to the existing structured records that exist within the police and intelligence organizations. |
|
|
| Insurance companies have the enviable position of having a rather predictable revenue stream (through insurance premiums), however the less attractive characteristic of highly variable operational costs (due to claims payouts). Therefore it is understandable that risk analysis plays a particularly important role in the average insurance company. The next question is: what can text analytics do to help? Similar to the police report example above, insurance claim information is highly unstructured. Most insurance companies attempt to apply some structure on the information gathered through forms-based systems used by their claims handlers. However, it is impossible to create a form to capture all the complex details that may affect liability and the sources of this real world information vary broadly from police officers, claims adjusters, hospital staff, interviews with involved parties and so on, all of which generate a lot of freeform text. Text Analytics is the only way that this information can contribute effectively to the decision making process and render a complete picture of risk. Currently this information relies on human readers and frequently gets overlooked. Make this information contribute to your bottom line instead of lying hidden at the bottom of a content management database. |
|
|
|
| | |
