Knowledge Search with Guided Conversation

When Luma Knowledge finds multiple relevant Artifacts that are related to different topics for a user request, a dynamic ‘Guided Conversation’ is generated to help the end-user find the required information. The end-user is presented with a series of questions or topics that the system uses to gather additional information to find the best matching Artifact. The end-user is intelligently guided to the most relevant information, instead of browsing through the list of artifacts. For more information, refer to Knowledge Search.

Following is an example of Luma Knowledge tenant for a City Municipal council where various Artifacts related to Permits are available in the Knowledge Store. The artifacts are associated with different topics such as Commercial Permit, Building Permit, etc. A multi-level hierarchy like the below is created by classifying the Artifacts under relevant Domains and Topics.

Open

Let's take an example search where the end-user is looking for information on applying for permits. The search results in a Guided conversation of topics to help the user find the most relevant knowledge artifact.

Search Widget

Knowledge Search in Search Widget is based on a Relevancy score. Consider the below example:

1. User searches for “How to apply for permit“.

2. Luma Knowledge identifies the following articles under various Topics and calculates respective relevancy scores:

   1. Topic: Permits, Licenses, and Inspections → Permit Center

      1. A1: "Permit Center Hours" with a Relevancy score of 6.57

   2. Topic: Permits, Licenses, and Inspections → Commercial Permits → "Building Permits"

      1. A1: "Permit Center Hours" with a Relevancy score of 6.57

      2. A2: "Multiple Housing Occupancy Permit" with a Relevancy score of 4.4

      3. A3: "Building Permit History" with a Relevancy score of 4.4

      4. A4: "Building Permits Online" with a Relevancy score of 4.4

   3. Topic: Permits, Licenses, and Inspections→ Commercial Permits → Signage

      1. A5: "Sign Ordinance" with a Relevancy score of 4.4

   4. Topic: COVIS

      1. A6: "Apply for COVIS access" with a Relevancy score of 6.4

3. Based on the following tenant configurations, results are displayed on the Search Widget:

All the Artifacts that fall in the Relevancy Score range, i.e., between the highest relevancy score for the matching artifacts (6.57) and the calculated Relevancy Threshold (3.94), are presented as the search result to the end-user.
Relevancy Threshold is calculated as Score Range Percentage (tenant configuration) of the highest relevancy score (60% of 6.57 = 3.94).

Open

The 6 Knowledge Artifacts identified as relevant are associated with different Topics. Based on the tenant configuration, Luma Knowledge disambiguates the relevant Topics and creates the decision levels. The following image defines the Artifact-Topic paths identified by the system.

Using the identified topic and decision levels, the system generates Guided conversation. Results are presented as shown below in Search Widget:

Luma Virtual Agent

When the Artifact is searched through Luma virtual Agent or any of the Luma support Channels, Knowledge is displayed based on the following factors:

Confidence Band

Knowledge Searches through Luma-supported Channels are governed by Confidence Score. The Confidence Score of an Artifact indicates how confident the system is on the artifact to help the user with their inquiry. The metadata generated from the user query is matched against the metadata of the Artifacts available in Knowledge Base. If at least one phrase/value in the metadata type is matched, the respective score is awarded to the Artifact as the Confidence score. For more details on calculating Confidence score, refer to Confidence Score Calculation.

In the current example,

1. User searches for “How to apply for permit“.

2. Luma Knowledge calculates the confidence score for the matched Artifacts (identified as per the relevancy score):

3. Based on the following tenant configurations, the matched Artifacts are classified in Confidence Bands. Consider that confidence thresholds for your tenant are as following:

Upper Confidence Score Threshold value for LOW Confidence level- 0.75
Upper Confidence Score Threshold value for MODERATE Confidence level - 0.5

The Artifacts are classified as:

The Highest Confidence Band of the identified Artifact is considered. For our search, the Knowledge Confidence Band is LOW.

User Intent

User intent is derived from the search phrase. It indicates if the user wants to view Knowledge or execute the predefined service. For the search “How to apply for permit“, user intent or pre-qualification is KNOWLEDGE.

Services on Luma Virtual Agent

The user’s search request is also used to search predefined services available in Luma VA. Based on the Luma tenant configurations, the Virtual Agent identifies matching services and calculates the skill(s) confidence level. Services are presented to the users along with the Knowledge Artifacts if matching services with High or Moderate Confidence Level are found. For more information on identifying Skills in Luma VA, refer to NLP Settings.

In the current example, let us consider that no pre-defined services are available in Luma VA. So, the Services’ Confidence level is LOW.

Result Presentation preference in Luma VA

Based on the above factors, the system recommends if the end-user should be presented with Knowledge, predefined Service, or both. For our search, the system recommends displaying Knowledge Followed by Skills. For more information on rules for Knowledge and Services display, refer to Integration with Luma Knowledge.

However, based on your organization’s representation preference, the Luma VA administrator can customize the way Knowledge and Services are delivered to the end-user. You can configure Luma VA to always deliver Knowledge followed by Skill, Knowledge & Service together or honor the DE recommendation. The configuration overrides the DE recommendation. For more information, Knowledge & Services Settings.

For the current example, let us consider that the Presentation preference for your Luma tenant is Default. This means the DE recommendation (Knowledge and Skills) is followed.

Now, based on the above factors and unavailability of matching Services, Knowledge Artifacts is presented to the end-user in Luma VA.

Since The 6 Knowledge Artifacts identified as relevant are associated with different Topics, the system disambiguates the relevant Topics and creates the decision levels. Luma VA presents the Guided conversation to disambiguate and find the relevant topic. Luma VA presents the results in a Guided conversation in the Virtual Agent, as shown below: