TAG REFINEMENT STRATEGIES FOR SOCIAL TAGGING SYSTEMS
Techniques are generally described related to tag refinement strategy. One example method for refining homonyms and synonyms in a plurality of tags may include receiving, by a tag refinement system, a plurality of tagging actions each of which associates one or more of the plurality of tags with a digital object. The method may further include extracting, by the tag refinement system, a first subset of tags from the plurality of tags, wherein the first subset of tags have a higher collective relativity-similarity score comparing to a second subset of tags selected from the plurality of tags, and the first subset of tags, different from the second subset of tags, have a same predetermined tag count as the second subset of tags.
1 . A method to generate a tag graph to represent a plurality of tags associated with a digital object, the method comprising:
connecting a first node and a second node with a first edge in the tag graph, wherein the first node represents a first tag, the second node represents a second tag, and the first edge represents a first one or more co-occurrence relationships between the first tag and the second tag;
assigning a first relativity score to the first node, wherein the first relativity score is based on a first number of one or more occurrences of the first tag;
assigning a second relativity score to the second node, wherein the second relativity score is based on a second number of one or more occurrences of the second tag; and
assigning a first similarity score to the first edge, wherein the first similarity score is based on a first number of the first one or more co-occurrence relationships between the first tag and the second tag.
2 . The method of claim 1 , wherein the first one or more co-occurrence relationships are created in response to annotating the digital object in each of one or more tagging actions with both the first tag and the second tag.
3 . The method of claim 1 , wherein the first number of the one or more occurrences of the first tag includes a first tag count of the first tag in the plurality of tags.
4 . The method of claim 1 , further comprising:
calculating a first relativity-similarity score for the first node based on the first relativity score, and a first one or more similarity scores of a first one or more edges of the first node.
5 . The method of claim 4 , further comprising:
determining if the first relativity-similarity score is higher than a first temporary relativity-similarity score; and
if the first relativity-similarity score is higher than the first temporary relativity-similarity score, selecting the first tag as a first one of a subset of tags.
6 . The method of claim 5 , further comprising:
setting the first relativity-similarity score as a second temporary relativity-similarity score.
7 . The method of claim 6 , further comprising:
calculating a second relativity-similarity score for the second node based on the second relativity score, and a second one or more similarity scores of a second one or more edges of the second node.
8 . The method of claim 7 , further comprising:
determining if the second relativity-similarity score is higher than the second temporary relativity-similarity score; and
if the second relativity-similarity score is higher than the second temporary relativity-similarity score, selecting the second tag as a second one of the subset of tags.
9 . The method of claim 8 , wherein the selecting the second tag as the second one of the subset of tags, includes:
replacing a tag from the subset of tags, wherein the replaced tag has a relativity-similarity score that is lower than second relativity-similarity score.
10 . The method of claim 1 , wherein the tag graph further includes:
a third node that represents a third tag, wherein the third tag is included in the plurality of tags; and
a second edge connecting the first node to the third node, wherein the second edge represents a second one or more co-occurrence relationships between the first tag and the third tag.
11 . The method of claim 10 , further comprising:
assigning a third relativity score to the third node, wherein the third relativity score is based on a third number of one or more occurrences of the third tag; and
assigning a second similarity score to the second edge, wherein the second similarity score is based on a second number of the second one or more co-occurrence relationships between the first tag and the third tag.
12 . A system configured to generate a tag graph to represent a plurality of tags associated with a digital object, the system comprising a processor and a memory coupled to the processor, wherein the memory is configured to provide the processor with instructions for:
connecting a first node and a second node with a first edge in the tag graph, wherein the first node represents a first tag, the second node represents a second tag, and the first edge represents a first one or more co-occurrence relationships between the first tag and the second tag;
assigning a first relativity score to the first node, wherein the first relativity score is based on a first number of one or more occurrences of the first tag;
assigning a second relativity score to the second node, wherein the second relativity score is based on a second number of one or more occurrences of the second tag; and
assigning a first similarity score to the first edge, wherein the first similarity score is based on a first number of the first one or more co-occurrence relationships between the first tag and the second tag.
13 . The system of claim 12 , wherein the first one or more co-occurrence relationships are created in response to annotating the digital object in each of one or more tagging actions with both the first tag and the second tag.
14 . The system of claim 12 , wherein the first number of the one or more occurrences of the first tag includes a first tag count of the first tag in the plurality of tags.
15 . The system of claim 12 , wherein the memory is configured to provide the processor with additional instructions for:
calculating a first relativity-similarity score for the first node based on the first relativity score, and a first one or more similarity scores of a first one or more edges of the first node.
16 . The system of claim 15 , wherein the memory is configured to provide the processor with additional instructions for:
determining if the first relativity-similarity score is higher than a first temporary relativity-similarity score; and
if the first relativity-similarity score is higher than the first temporary relativity-similarity score, selecting the first tag as a first one of a subset of tags.
17 . The system of claim 16 , wherein the memory is configured to provide the processor with additional instructions for:
setting the first relativity-similarity score as a second temporary relativity-similarity score.
18 . The system of claim 17 , wherein the memory is configured to provide the processor with additional instructions for:
calculating a second relativity-similarity score for the second node based on the second relativity score, and a second one or more similarity scores of a second one or more edges of the second node.
19 . The system of claim 18 , wherein the memory is configured to provide the processor with additional instructions for:
determining if the second relativity-similarity score is higher than the second temporary relativity-similarity score; and
if the second relativity-similarity score is higher than the second temporary relativity-similarity score, selecting the second tag as a second one of the subset of tags.
20 . The system of claim 19 , wherein the selecting the second tag as the second one of the subset of tags, includes:
replacing a tag from the subset of tags, wherein the replaced tag has a relativity-similarity score that is lower than second relativity-similarity score.
21 . The system of claim 12 , wherein the tag graph further includes:
a third node that represents a third tag, wherein the third tag is included in the plurality of tags; and
a second edge connecting the first node to the third node, wherein the second edge represents a second one or more co-occurrence relationships between the first tag and the third tag.
22 . The system of claim 21 , wherein the memory is configured to provide the processor with additional instructions for:
assigning a third relativity score to the third node, wherein the third relativity score is based on a third number of one or more occurrences of the third tag; and
assigning a second similarity score to the second edge, wherein the second similarity score is based on a second number of the second one or more co-occurrence relationships between the first tag and the third tag.