System and method for user-oriented topic selection and browsing

A non-transitory computer readable medium having stored therein a program for making a computer execute a method of displaying a plurality of content items, the method comprising:

• detecting at least one topic associated with each of the plurality of content items, each of the at least one topic characterized by topic terms;
• generating a set of topics based on the at least one topic associated with each of the plurality of content items, the set of topics comprising a number of topics associated with the plurality of content items;
• for each respective topic of the set of topics, calculating similarity scores between topic terms of each respective topic and topic terms of one or more other topics in the set of topics, wherein the calculating the similarity scores includes:
• calculating a first similarity score between topic terms associated with a first topic in the set of topics and topic terms associated with a second topic in the set of topics;
• calculating a second similarity score between the topic terms associated with the second topic and topic terms associated with a third topic in the set of topics;
• and
• calculating a third similarity score between the topic terms associated with the first topic and the topic terms associated with the third topic;
• reducing the number of topics in the set of topics by combining at least one topic with the one or more other topics in the set of topics in response to a determination that the at least one topic is redundant based, in part, on the calculated similarity scores between topic terms of the at least one topic and the one or more other topics exceeding a first threshold, wherein the first threshold is a non-zero threshold, and wherein the reducing the number of topics includes:
• determining the first topic, the second topic, and the third topic are redundant when each of the first similarity score, the second similarity score, and the third similarity score exceed the first threshold;
• calculating a first total similarity value associated with the first topic based on a summation of the first similarity score and the third similarity score;
• calculating a second total similarity value associate with the second topic based on a summation of the first similarity score and the second similarity score;
• calculating a third total similarity value associated with the third topic based on a summation of the second similarity score and the third similarity score;
• determining a largest similarity value from the first total similarity value, the second total similarity value, and the third total similarity value;
• selecting a representative topic from the first topic, the second topic, and the third topic based on the determination of the largest similarity value;
• and
• determining that a topic in the set of topics is non-redundant based, in part, on the calculated similarity scores between the topic and the one or more other topics in the set of topics is less than the first threshold;
• calculating a topic distribution for each of the plurality of content items from the reduced set of topics;
• and
• generating a visualization, the visualization comprising
• a first region displaying visual representations of each topic in the reduced set of topics, each visual representation of each topic comprising a topic order indicator;
• and
• a second region displaying visual representations of a subset of content items of the plurality of content items, each visual representation of the subset of content items comprising a plurality of topic indicators that each correspond to a topic order indicator of the first region, each topic indicator comprising a graphical representation of the topic distribution of the corresponding topic in first region for a respective content item in the second region.

A computing device configured to display a plurality of content items, the computing device comprising:

• a memory configured to store a plurality of content items;
• a processor executing a process comprising:
• detecting at least one topic associated with each of the plurality of content items, each of the at least one topic characterized by topic terms;
• generating a set of topics based on the at least one topic associated with each of the plurality of content items, the set of topics comprising a number of topics associated with the plurality of content items;
• for each respective topic of the set of topics, calculating similarity scores between topic terms of each respective topic and topic terms of one or more other topics in the set of topics, wherein the calculating the similarity scores includes:
• calculating a first similarity score between topic terms associated with a first topic in the set of topics and topic terms associated with a second topic in the set of topics;
• calculating a second similarity score between the topic terms associated with the second topic and topic terms associated with a third topic in the set of topics;
• and
• calculating a third similarity score between the topic terms associated with the first topic and the topic terms associated with the third topic;
• reducing the number of topics in the set of topics by combining at least one topic with the one or more other topics in the set of topics in response to a determination that the at least one topic is redundant based, in part, on the calculated similarity scores between topic terms of the at least one topic and the one or more other topics exceeding a first threshold, wherein the first threshold is a non-zero threshold, and wherein the reducing the number of topics includes:
• determining the first topic, the second topic, and the third topic are redundant when each of the first similarity score, the second similarity score, and the third similarity score exceed the first threshold;
• calculating a first total similarity value associated with the first topic based on a summation of the first similarity score and the third similarity score;
• calculating a second total similarity value associate with the second topic based on a summation of the first similarity score and the second similarity score;
• calculating a third total similarity value associated with the third topic based on a summation of the second similarity score and the third similarity score;
• determining a largest similarity value from the first total similarity value, the second total similarity value, and the third total similarity value;
• selecting a representative topic from the first topic, the second topic, and the third topic based on the determination of the largest similarity value;
• and
• determining that a topic in the set of topics is non-redundant based, in part, on the calculated similarity scores between the topic and the one or more other topics in the set of topics is less than the first threshold;
• calculating a topic distribution for each of the plurality of content items from the reduced set of topics;
• and
• generating a visualization, the visualization comprising
• a first region displaying visual representations of each topic in the reduced set of topics, each visual representation of each topic comprising a topic order indicator;
• and
• a second region displaying visual representations of a subset of content items of the plurality of content items, each visual representation of the subset of content items comprising a plurality of topic indicators that each correspond to a topic order indicator of the first region, each topic indicator comprising a graphical representation of the topic distribution of the corresponding topic in first region for a respective content item in the second region.