Method of and system for inferring user intent in search input in a conversational interaction system

A computer-implemented method of inferring user intent in a search input, the method comprising:

• receiving a search input from a user at one or more processors, the search input used to identify at least one desired content item of a set of content items, each of the content items being associated with at least one domain;
• determining by the one or more processors that a first portion of the search input contains a linguistic element that has a plurality of possible meanings;
• calculating a first weight based on a distance between a first node of a first graph data structure corresponding to the first portion of the search input and a second node of the first graph data structure corresponding to a second portion of the search input, wherein the first graph data structure comprises nodes connected by relationships, and wherein the first graph data structure is relevant to a first domain;
• calculating a second weight based on a distance between a first node of a second graph data structure corresponding to the first portion of the search input and a second node of the second graph data structure corresponding to the second portion of the search input, wherein the second graph data structure comprises nodes connected by relationships, and wherein the second graph data structure is relevant to a second domain;
• determining by the one or more processors one of the first domain or the second domain as being a relevant domain to the search input based on a comparison of the first weight and the second weight;
• selecting by the one or more processors at least one content item from the set of content items based on the search input, the relevant domain, and metadata associated with the content items;
• and
• causing to be provided information corresponding to the at least one desired content item.