Method for generating numerical values indicative of frequencies of selected features in objects, and a computer system implementing the method

A computer system, comprising:

• a memory;
• a central processing unit (CPU) coupled to the memory and configurable to obtain program instructions from the memory, and to execute the program instructions;
• wherein during operation of the computer system, program instructions stored in the memory cause the CPU to:
• arrange columns of a matrix in sum total order, wherein the matrix has at least one row and a plurality of intersecting columns, wherein each of the rows represents a different one of a plurality of objects, and wherein each of the columns represents a different one of a plurality of selected features, and wherein values reside at row-column intersections, and wherein a given value residing at an intersection of a given row and a given column corresponds to the given row and the given column, and represents the frequency of the selected feature represented by the given column in the object represented by the given row;
• convert the matrix to a binary matrix comprising binary values ‘0’ and ‘1’;
• divide the columns of the matrix into a plurality of segments of equal length;
• replace matrix columns encompassed by each segment by a single column, and setting values at intersections of the rows and the single columns equal to numerical values indicative of ratios of a total number of one of the binary values in a portion of the corresponding row encompassed by a segment to a total number of the one of the binary values in the corresponding row;
• form a plurality of text strings using the values at the intersections of the rows and the single columns;
• and
• associate each of the plurality of text strings with the object represented by the row corresponding to the value used to form the text string.

A computer system, comprising:

• a memory, wherein during operation of the computer system, program instructions are stored in the memory operable to implement:
• arranging columns of a matrix in sum total order, wherein the matrix has at least one row and a plurality of intersecting columns, wherein each of the rows represents a different one of a plurality of objects, and wherein each of the columns represents a different one of a plurality of selected features, and wherein values reside at row-column intersections, and wherein a given value residing at an intersection of a given row and a given column corresponds to the given row and the given column, and represents the frequency of the selected feature represented by the given column in the object represented by the given row;
• converting the matrix to a binary matrix comprising binary values;
• dividing the columns of the matrix into a plurality of segments of equal length;
• replacing matrix columns encompassed by each segment by a single column, and setting values at intersections of the rows and the single columns equal to numerical values indicative of ratios of a total number of one of the binary values in a portion of the corresponding row encompassed by a segment to a total number of the one of the binary values in the corresponding row;
• forming a plurality of text strings using the values at the intersections of the rows and the single columns;
• associating each of the plurality of text strings with the object represented by the row corresponding to the value used to form the text string;
• receiving a query;
• and
• selecting at least one of the plurality of objects based upon the query and the plurality of text strings;
• and
• a central processing unit (CPU) coupled to the memory and configurable to obtain the program instructions from the memory, and to execute the program instructions.

A computerized method, comprising:

• determining a frequency of each of a plurality of selected features in each of at least one object;
• using the frequency information to construct a two-dimensional matrix having at least one row and a plurality of intersecting columns, wherein each of the rows represents a different one of the at least one objects, and wherein each of the columns represents a different one of the selected features, and wherein values reside at row-column intersections, and wherein a given value residing at an intersection of a given row and a given column corresponds to the given row and the given column, and represents the frequency of the selected feature represented by the given column in to object represented by the given row;
• summing the values in each of the columns of the matrix to produce a sum total for each of the columns;
• arranging the columns of the matrix in sum total order;
• converting the matrix to a binary matrix comprising values of ‘0’ and ‘1’;
• dividing the columns of the matrix into a plurality of segments of equal length;
• and
• replacing matrix columns encompassed by each segment by a single column, and setting values at intersections of the rows and the single columns equal to rational fractions formed by dividing a total number of ‘1’ values in a portion of the corresponding row encompassed by a segment by a total number of ‘1’ values in the corresponding row;
• forming a plurality of text strings using the values at the intersections of the rows and the single columns;
• associating each of the plurality of text strings with the object represented by the row corresponding to the value used to form the text string;
• receiving a query;
• and
• selecting at least one of the plurality of objects based upon the query and the plurality of text strings.

A carrier medium comprising program instructions, wherein the program instructions are operable to implement:

• arranging columns of a matrix in sum total order, wherein the matrix has at least one row and a plurality of intersecting columns, wherein each of the rows represents a different one of a plurality of objects, and wherein each of the columns represents a different one of a plurality of selected features, and wherein values reside at row-column intersections, and wherein a given value residing at an intersection of a given row and a given column corresponds ta the given row and the given column, and represents the frequency of the selected feature represented by the given column in the object represented by the given row;
• converting the matrix to a binary matrix comprising binary values;
• dividing the columns of the matrix into a plurality of segments of equal length;
• and
• replacing matrix columns encompassed by each segment by a single column, and setting values at intersections of the rows and the single columns equal to numerical values indicative of ratios of a total number of one of the binary values in a portion of the corresponding row encompassed by a segment to a total number of the one of the binary values in the corresponding row;
• forming a plurality of text strings using the values at the intersections of the rows and the single columns;
• and
• associating each of the plurality of text strings with the object represented by the row corresponding to the value used to form the text string.