SYSTEMS AND METHODS FOR AUTOMATIC GENERATION OF A DOMAIN TAXONOMY
A computerized system and method may automatically generate a hierarchical, multi-tiered taxonomy based on measuring and/or quantifying degrees of generality for a plurality of input entities. In some embodiments of the invention, a computerized system comprising a processor, and a memory including a plurality of entities such as documents or text files, may be used for extracting words from a plurality of documents; calculating generality scores for the extracted words; selecting some of the extracted words as exemplars based on the scores; and clustering unselected words under appropriate exemplars to produce or output a corresponding taxonomy. Some embodiments of the invention may allow categorizing interactions among remotely connected computers using a domain taxonomy, and routing interactions between remotely connected computer systems based on the taxonomy.
1 . A method for automatic generation of a domain taxonomy, the method comprising:
in a computerized-system comprising a processor, and a memory including a plurality of entities:
automatically generating, by the processor, the domain taxonomy, the generating comprising:
(i) calculating, by the processor, a generality score for one or more nodes, each node comprising one of the entities or a cluster of entities;
(ii) selecting, by the processor, one or more of the nodes as exemplars based on the calculated scores; and
(iii) clustering, by the processor, one or more unselected nodes under one or more of the exemplars.
2 . The method of claim 1 , wherein the calculating of a generality score comprises calculating a frequency of occurrence for one or more of the entities.
3 . The method of claim 1 , wherein the calculating of a generality score comprises, for a given entity, identifying one or more of the entities as joint-entities based on at least one of: a distance from the given entity, and being linked to the given entity by a dependency parser.
4 . The method of claim 3 , wherein the calculating of a generality score comprises calculating a joint-entity-spread index based on a distance of each joint-entity from the given entity.
5 . The method of claim 2 , wherein the calculating of a generality score comprises calculating a weighted frequency of occurrence for a given entity based on frequencies of occurrence of one or more other entities.
6 . The method of claim 1 , comprising:
calculating, by a vector embedding model, one or more vector representations for one or more of the nodes;
receiving one or more additional entities and clusters; and
clustering, by the model, one or more of the additional entities and clusters based on the calculated vector representations.
7 . The method of claim 1 , comprising:
Providing, by the processor, a plurality of search results for an input query based on the taxonomy.
8 . The method of claim 1 , wherein one or more of the entities include one or more words extracted from one or more documents.
9 . A computerized system for automatic generation of a domain taxonomy, the system comprising:
a computer processor,
and a memory including a plurality of entities;
wherein the processor is configured to automatically generate the domain taxonomy, the generating comprising:
(i) calculating a generality score for one or more nodes, each node comprising one of the entities or a cluster of entities;
(ii) selecting one or more of the nodes as exemplars based on the calculated scores; and
(iii) clustering one or more unselected nodes under one or more of the exemplars.
10 . The computerized system of claim 9 , wherein the calculating of a generality score comprises calculating a frequency of occurrence for one or more of the entities.
11 . The computerized system of claim 9 , wherein the calculating of a generality score comprises, for a given entity, identifying one or more of the entities as joint-entities based on at least one of: a distance from the given entity, and being linked to the given entity by a dependency parser.
12 . The computerized system of claim 11 , wherein the calculating of a generality score comprises calculating a joint-entity-spread index based on a distance of each joint-entity from the given entity.
13 . The computerized system of claim 10 , wherein the calculating of a generality score comprises calculating a weighted frequency of occurrence for a given entity based on frequencies of occurrence of one or more other entities.
14 . The computerized system of claim 9 , wherein the processor is configured to:
calculate, by a vector embedding model, one or more vector representations for one or more of the nodes;
receive one or more additional entities and clusters; and
cluster, by the model, one or more of the additional entities and clusters based on the calculated vector representations.
15 . The computerized system of claim 9 , wherein the processor is configured to provide a plurality of search results for an input query based on the taxonomy.
16 . The computerized system of claim 9 , wherein the memory includes one or more documents, and wherein one or more of the entities include one or more words extracted from one or more of the documents.
17 . A method for categorizing interactions using an automatically generated domain taxonomy, the method comprising:
in a computerized-system comprising a processor, and a memory including a data store of a plurality of documents, and connected by a network to one or more remote computers:
automatically generating, by the processor, the domain taxonomy based on the one or more documents, the generating comprising:
(i) extracting a plurality of words from the documents;
(ii) calculating, by the processor, a generality score for one or more nodes, each node comprising one or more of the words or a cluster of words;
(iii) selecting, by the processor, one or more of the nodes as exemplars based on the calculated scores;
(iv) clustering, by the processor, one or more unselected nodes under one or more of the exemplars; and
(v) iteratively repeating steps (ii)-(iv) until one or more convergence criteria are satisfied; and
categorizing, by the processor, one or more interactions routed to one or more of the remote computers, wherein the categorizing is performed using the generated taxonomy.
18 . The method of claim 17 , wherein one or more of the documents describe one or more interactions, the interactions routed using a private branch exchange to one or more of the remote computers.
19 . The method of claim 18 , comprising: routing, by the private branch exchange, one or more of the interactions to a remote computer among the one or more remote computers based on the taxonomy.
20 . The method of claim 17 , wherein the calculating of a generality score comprises calculating a frequency of occurrence for one or more of the entities.