About this Module
The Recommendation is that you can set up the Business Matchmaking matrix to connect your participants - attendees, speakers, sponsors/exhibitors and propose conference sessions of their interests.
This Module is an “inlet” through which the Matrix is added. After that, you can create a Recommendation block in the livepages to display the Recommendations.
If this page is for Matrix settings, where can I enable Recommendations for the event? → Recommendations under the LiveBar If I do not set any Rules, can I still use the Recommendation Features? → Yes, the engine will learn from the participant's behaviours within the siites and suggest relevant leads and session to them.
Clicking on the Add Rule button gives you a pop-up on the right.
The priority that you would like to give to this rule
Rule with priority 1 will have the highest priority/importance/weightage and will be lower as the priority field value increases. Rules can also have equal priorities.
This option allows us to create 3 different criteria for recommendations:
- Inclusion - A normal recommendation rule
- Exclusion - A rule to exclude a particular set of attendees to be recommended to or to get recommendations of some other particular set of attendees.
- ExcludeAll - A rule to completely exclude a set of Attendees from recommendations
Note: Criteria are available only for the PEOPLE TO PEOPLE type.
Groups can be created in cases where it is required to apply single or multiple rules to a particular group of Attendees.
This can be any one of the fields, available to people on the registration forms.
Example: If you select Job Title, it will use the values from the Job Title field as a source to match with that of the target field.
If you want to recommend some specific source value to some specific target value then you can specify the source value in this text field.
Leave this text field empty if you want all possible source values to be used for comparison.
- Exact Match:
Also known as Syntactic Match. This type will generate recommendations for every exact match between source and target values.
- Semantic Match:
In general, semantics means the meaning of a word or a sentence. In this type, the meaning of the text in the source and target values will be compared using NLP.
The target field values are compared with the source field values to generate recommendations.
Example: In the case of the People to People type of recommendation you can select any people fields. Similarly in the case of People to Session or People to Sponsor type of recommendations you can select any field from sessions or sponsors for recommendations respectively with the most common ones, for both, being name, description, tags, etc.
Similar to Field Value, this can be used in situations where you want to recommend some specific source value to some specific target value. You can specify the target value in this text field.
Leave this text field empty if you want all possible target values to be used for comparison.
Two Way Matching
In Rules where Source Field and Target Fields are specified only the Target will be recommended to the Source. This Two-Way-Matching allows us to recommend Source to Target in addition to Target to Source.
Note: Two Way Matching is available only for the PEOPLE TO PEOPLE type.
In this example, we define a rule named Similar Job Title for People to People type of recommendation. The Source and target fields are jobTitle and their corresponding Field values are left empty and the type is Exact Match.
This will generate attendee recommendations for people who have similar job titles.
Example: Attendees having similar job titles will be recommended to each other.
This rule will generate session recommendations for people whose job title semantically matches the session's description.
Example: An attendee with a job title Machine Learning Engineer will get recommendations of Sessions having descriptions related to Machine Learning. An example of such a session description is given below:
“In this session our keynote speaker Mr. Dustin Porier who works as an AI Engineer in UFC will be explaining about the entire machine learning life cycle with its implementation using AWS Sagemaker”
This rule generates recommendations of people working in Nokia Company for Microsoft Employees and vice-versa.
Since the field value and target value text fields are populated with the company Nokia and Microsoft respectively, attendees who work in Nokia will be recommended to attendees working in Microsoft and vice versa.
This rule will generate session recommendations for people whose Interests field values match with the session tags.
Example: An attendee with an Interest in Economics will get recommendations of Sessions having a tag Economics.