Ranking proximity of data sources with authoritative entities in social networks

A computerized system for ranking a proximity of data sources with one or more reference entities, the system comprising:

• a memory storing computerized methods and identifiers of the data sources and the one or more reference entities;
• and
• one or more processing elements;
• wherein the computerized methods are designed, when executed by said one or more processing elements, to:
• obtain, for each identified reference entity, a vector of terms representative of said each reference entity;
• determine a measure of information provided by the terms in the vector obtained for said each reference entity, wherein said measure of information is determined, for each term w, by computing an inverse document frequency IDF as IDF=1+log(M/(1+N(w), where M is the total number of arrays as determined for all the identified data sources and N is the number of arrays that comprise said each term;
• for each data source of the identified data sources, select terms from said each data source and populate an array comprising the terms;
• determine, in said array, a term frequency of terms of the vector obtained for said each of the reference entities, wherein the term frequency corresponds to the number of occurrences of said each of the terms in said array;
• weight each term frequency, as determined for each data source, by the measure of information as determined for a term corresponding to a term for which said each term frequency was determined, to obtain a weighted term frequency for the term, and
• rank a proximity of the identified data sources with each of the one or more reference entities, according to the weighted term frequencies as determined for all the terms of the vector obtained for said each reference entity.

A computer-program product for ranking a proximity of data sources with one or more reference entities, the computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions being executable by one or more processing elements to cause the latter to:

• identify the data sources and the one or more reference entities;
• obtain, for each identified reference entity, a vector of terms representative of said each reference entity;
• determine a measure of information provided by the terms in the vector obtained for said each reference entity, wherein said measure of information is determined, for each term w, by computing an inverse document frequency IDF as IDF=1+log(M/(1+N(w), where M is the total number of arrays as determined for all the identified data sources and N is the number of arrays that comprise said each term;
• for each data source of the identified data sources, select terms from said each data source and populate an array comprising the selected terms;
• determine, in said array, a term frequency of terms of the vector obtained for said each of the reference entities, wherein the term frequency corresponds to the number of occurrences of said each of the terms in said array;
• weight each term frequency, as determined for each data source, by the measure of information as determined for a term corresponding to a term for which said each term frequency was determined, to obtain a weighted term frequency for the term;
• rank a proximity of the identified data sources with each of the one or more reference entities, according to the weighted term frequencies as determined for all the terms of the vector obtained for said each reference entity.