Methods, systems, and computer readable media for dynamic cluster-based search and retrieval

A system for dynamic cluster-based search and retrieval, the system comprising:

• a computing platform including at least one processor:
• a search and retrieval application executable by the at least one processor for:
• retrieving document data for a plurality of documents related to user input:
• performing keyword discovery on the document data for determining term related frequency metrics and document related frequency metrics:
• representing the plurality of documents as a term-document matrix based on the term related frequency metrics and the document related frequency metrics:
• reducing, using latent semantic analysis, the dimensionality of the matrix:
• clustering, using a sampling-based k-means clustering algorithm and the dimensionally reduced matrix, the plurality of documents into clusters, wherein the plurality of documents is a predetermined number of most recent documents related to the user input, wherein the most recent documents are determined using document dates in a data store, wherein the sampling-based k-means clustering algorithm executes in constant or near constant time;
• and
• sending presentation information to a client device for displaying visual representations of the clusters, wherein each of the visual representations is associated with one or more of the plurality of documents, wherein sending the presentation information includes sending related keywords for the clusters and spatial information for the clusters, and wherein the related keywords for the clusters are determined by:
• for each of the clusters:
• identifying a centroid of the cluster;
• transforming the centroid to a term vector containing term frequency information;
• and
• determining the related keywords based on terms with the highest frequency as indicated by the term vector.

A non-transitory computer readable medium having stored thereon executable instructions that when executed by at least one processor of a computer cause the computer to perform steps comprising:

• al a server retrieving document data for a plurality of documents related to user input:
• performing keyword discovery on the document data for determining term related frequency metrics and document related frequency metrics:
• representing the plurality of documents as a term-document matrix based on the term related frequency metrics and the document related frequency metrics:
• reducing, using latent semantic analysis, the dimensionality of the matrix:
• clustering, using a sampling-based k-means clustering algorithm and the dimensionally reduced matrix, the plurality of documents into clusters, wherein the plurality of documents is a predetermined number of most recent documents related to the user input, wherein the most recent documents are determined using document dates in a data store, wherein the sampling-based k-means clustering algorithm executes in constant or near constant time;
• and
• sending presentation information to a client device for displaying visual representations of the clusters, wherein each of the visual representations is associated with one or more of the plurality of documents, wherein sending the presentation information includes sending related keywords for the clusters and spatial information for the clusters, and wherein the related keywords for the clusters are determined by:
• for each of the clusters:
• identifying a centroid of the cluster;
• transforming the centroid to a term vector containing term frequency information;
• and
• determining the related keywords based on terms with the highest frequency as indicated by the term vector.