Knowledge Discovery Core Components
This section describes the common Knowledge Discovery components.
The following components form the core text indexing and processing functions. These components are all ACI servers. For more information, refer to the Knowledge Discovery Help.
Service | Description |
---|---|
Content![]() |
Indexes, collects, manipulates and stores unstructured and semi-structured text data. |
AgentStore![]() |
Indexes, collects, manipulates and stores agent and category information. Agentstore is a special configuration of the Content component. |
Category |
Creates and manages categories, and performs text analysis such as clustering and taxonomy generation. |
Community![]() |
Stores information about users, roles, and permissions. |
View![]() |
Converts documents to HTML format for viewing in a Web browser. |
The following components form part of a wider Knowledge Discovery infrastructure. They allow you to license, distribute, monitor, and control your components. These components are also all ACI servers.
Service | Description |
---|---|
License Server![]() |
Distributes and manages licenses for Knowledge Discovery products and components. Refer to the License Server Help for more information. |
Proxy![]() |
An optional component that can route actions to the correct components. |
DAH ![]() |
Distributed Action Handler. Distributes action requests across Knowledge Discovery components. Refer to the DAH Help for more information. |
DIH ![]() |
Distributed Index Handler. Distributes indexing requests across Content components. Refer to the DIH Help for more information. |
Controller![]() |
Monitors the Knowledge Discovery services on a host machine, and communicates with Coordinator to report status information. This component is primarily used by Site Admin. Refer to the Site AdminUser Guide for more information. |
Coordinator![]() |
Manages status information from Controllers, and acts as a central point for viewing logs and monitoring the status of the Knowledge Discovery system. This component is primarily used by Site Admin. Refer to the Site Admin User Guide for more information. |
The following components allow you to automatically retrieve content and index into Knowledge Discovery.
Service | Description |
---|---|
NiFi Ingest![]() |
Allows you to easily configure and manipulate your data ingest process, from your connectors, to File Content Extraction and other import processes (such as media analysis and Eduction), and your Knowledge Discovery index. It has a user interface based on Apache NiFi, which you use to configure your whole ingest chain. You can use it to retrieve data from many repositories, use data enrichment (with routing to ensure different file types get appropriate processing), and index into the Content component, or update data in your repositories. Refer to the NiFi Ingest Help for more information. |
CFS![]() |
Connector Framework Server. Aggregates data from connectors and indexes the data into the Content component. CFS can also manipulate the documents sent to it by the connector (for example, to add additional metadata fields generated by data enrichment tasks). CFS is an ACI server. Refer to the Connector Framework Server Help for more information. |
Connectors![]() |
Gather data from different sources for indexing into the Content component. In some cases, connectors can also insert back into the original repositories. There are connectors for many different repositories. For NiFi based system, each connector is a NiFi processor, which you configure from the NiFi interface. For CFS-based systems, each connector is an ACI server, which you configure separately. Refer to the NiFi Ingest Help or the Connector Administration guides for more details. |
The following components perform additional operations and analysis on unstructured data. You can use many of these components with Knowledge Discovery text processing, or as part of an independent system. These components are all available as ACI servers, and some are also available in other formats.
Service | Description |
---|---|
Media Server![]() |
Media Server analyzes video files and streams, images, and audio to extract information about their content. It can run analysis operations such as face recognition, number plate recognition, speech-to-text, and speaker identification. Media Server is an ACI Server. A lot of Media Server functionality is also available in the NiFi media analysis processors, which you can use with NiFi Ingest. Refer to the Media Server Help for more information. |
Eduction |
Extracts entities (small pieces of information) from text, such as names, addresses, and telephone numbers. Eduction is available as an ACI server to use with other Knowledge Discovery services, as a separate SDK, or embedded in NiFi Ingest and CFS. Refer to the Eduction User and Programming Guide and the Eduction Server Reference for more information. |
Answer Server |
Performs Search and Retrieval and Conversations. Answer Server can answer questions from a database of facts, from a stored list of questions such as an FAQ, or by returning a relevant passage from a relevant store of unstructured data. It can also manage ongoing conversations with context, for example to run a chatbot. Refer to the Answer Server Help for more information. |
QMS ![]() |
Query Manipulation Server. Manages promotions, modifies queries to the Content component, and manipulates results from the Content component. For example, you can use QMS to remove certain terms from the query text, set up promotions to return when a user searches for a particular item, or to return a promoted item to a particular position in the search results. Refer to the Query Manipulation Server Help for more information. |
Statistics Server![]() |
Accumulates events from client applications or from a script that reads log files, then uses that data to report statistics. Refer to the Query Manipulation Server Help for more information. |
OGS![]() |
OmniGroupServer. Collects security information from security repositories and uses it to ensure that users can access only documents that they have permission to access. Refer to the Document Security Administration Guide and OmniGroupServer Reference for more information. |
Knowledge Graph![]() |
Uses connections between documents to create a set of relationships called a graph, and allows you to explore these connections in your data. Refer to the Knowledge Graph Help for more information. |