|Bookshelf Home | Contents | Index | PDF|
Smart Answer Manager is the server component used by Siebel Smart Answer to detect the meaning or intent of incoming free-form email inquiries for eMail Response, and specified texts from inquiries for the employee applications, such as Call Center and customer applications, such as eService.
Smart Answer Manager uses natural language processing (NLP) and semantic analysis to interpret the meaning and intent of email messages, generate concept models, and assign a confidence level percentage.
After Smart Answer receives a text entry, which could be in the form of an email or an entry in the eService Ask a Question applet, the statistical processing engine compares the concepts extracted from the message to the concept model for each of the categories. The NLP engine then categorizes the emails based on this concept model comparison, allowing Smart Answer to pass categories with a confidence-level score to Siebel Applications, such as eMail Response or eService. This percentage represents the probability that the text falls within that category.
If the confidence level of a category's accuracy is high, you may want eMail Response to automatically respond directly to the sender. If the confidence level is low, Smart Answer may suggest responses to the agent for approval.
For Call Center, Smart Answer will take the text from Service Request Summary and send this to Smart Answer for categorization when the agent clicks the Smart Answer Button. This opens up Search Center for Smart Answer and a list of content types are presented. The agent has the ability to send feedback to the system by clicking the Was this information helpful button.
A KB file stores information accumulated during training and ongoing learning, allowing Smart Answer Manager to accurately categorize similar incoming requests and queries that are received and processed. A flat KB file contains one level of categories and is easier to create and maintain than a hierarchical KB file. A hierarchical KB file uses a hierarchy of categories. Multiple levels of categories may be beneficial when similar concepts appear in more than one category. By grouping categories under a branch, it is easier to locate a specific template associated with it.
|Siebel Smart Answer Administration Guide|