Method for sorting a defined set of comments

Description

FIELD OF THE INVENTION

This invention relates to method for sorting a defined set of comments.

BACKGROUND OF THE INVENTION

Business and individuals have been interested in comments for many years. For the past few years, comments from blogs, social media, surveys and other media have become increasingly important. These comments have traditionally been sorted chronologically or read by individuals to determine meaning and determine the most important or useful comments. The present invention provides a method for sorting the comments into a more meaningful order than chronological and less cumbersome than manual reading.

BRIEF SUMMARY OF THE INVENTION

The present invention includes methods, systems and other means of evaluating and ranking comments from any source. The method for sorting a defined set of comments comprises: first, calculating a normalized word score. Then using the normalized word score to calculate a unique word score. After that the method uses the raw comment score to calculate a normalized comment score; and finally the method sorts the defined set of comments by the normalized comment score.

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

Having thus described the invention in general terms, reference will now be made to the accompanying drawings, which are not necessarily drawn to scale, and wherein:

FIG. 1 shows a schematic of the method involved in the present invention

DETAILED DESCRIPTION OF THE INVENTION

Embodiments of the present invention overcome many of the obstacles associated with discerning between impacting and non-impacting comments, and now will be described more fully hereinafter with reference to the accompanying drawings that show some, but not all embodiments of the claimed inventions. Indeed, the invention may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will satisfy applicable legal requirements.

FIG. 1 shows the method of the invention. The method starts with a user posting some media on a website to which readers can comment. After that, the website gathers comments, responses and answers from readers from any source. Beyond merely commenting on the website itself, the website will trackback links to other websites where the media is being discussed. Each comment has a total number of words (Tw) and a total number of raw words (Rw). A raw word is a particular word that can be any noun, verb, adjective, adverb, acronym, abbreviation or other series of contiguous letters. Words such as articles (a, an, the) and prepositions are removed automatically, other words can be removed manually as explained below.

The first criteria measurement in the method is to calculate the raw word score. The raw word score (Rs) is the number of times a particular word appears in all of the comments, responses and answers collectively. In some embodiments, the raw word score (Rs) is the number of times a particular word appears in all of the comments, responses and answers collectively plus an additional amount for the number of times a particular word appears in all of the comments, responses and answers that generate additional activity such as additional comments, repasts, shares or likes.

The second criteria measurement in the method is to calculate the normalized words score (Ns). Of all the words in the first criteria, the base number (Bn) is the largest raw word score. Normalized word score is raw word score divided by base number times one hundred. Therefore, each word has its score normalized by the following formula:

Ns ⁢ ⌈ Rs Bn * 10 0 ⌉ eqn . ⁢ 1

The third criteria measurement in the method is to calculate the unique words score (Uswn). In order to ensure that each word has a unique score, the words are sorted by Ns and then alphabetically. The unique word score for words with a unique normalized word score equals the unique normalized word score. Therefore, where there is only one word with an Ns; then
Ns=Uswn  eqn. 2

as the score is already unique.

Where there is more than one word with an identical Ns, then the Uswn for the nth word alphabetically is
Uswn=Ns+0.00001×n  eqn. 3

The fourth criteria measurement in the method is to calculate a raw comment score (RCs) for a particular comment, response or answer. The raw comment score is a sum of unique word scores for each word in that comment.

RCs = ∑ s n ⁢ Yswn eqn . ⁢ 4

The fifth criteria measurement in the method is to calculate a normalized comment score (NCs). Of all the comments in the fourth criteria, the base comment number (BCn) is the raw comment score of the comment which has the greatest raw comment score. The normalized comment score is the normalized comment score is the raw comment score divided by a base comment score times one hundred. Therefore, each comment has its score normalized by the following formula:

NCs = ⌈ RCs BCn * 10 0 ⌉ eqn . ⁢ 5

The sixth criteria measurement in the method is to calculate an average raw comment score (ARCs). The average raw comment score is equal to the raw comment score divided by the total number of words in the comment (Tw).

ARCs = RCs Tw eqn . ⁢ 6

The seventh criteria measurement in the method is to calculate a normalized average raw comment score (NARCs). Of all the comments in the sixth criteria, the base average comment number (BACn) is the raw comment score of the comment which has the greatest raw comment score. The normalized average raw comment score is the normalized average raw comment score is the average raw comment score, divided by the base average comment number, times one hundred. Therefore, each comment has its score normalized by the following formula:

NARCs = ⌈ ARCs BACn * 10 0 ⌉ eqn . ⁢ 7

At this point the user has a choice as to which of the previous seven measurements the user wants to utilize to sort the comments, answers and responses. It is most effective to use either the normalized comment score or the normalized average comment score.

If the user selects the normalized comment score then a list is produced which has grouped the comments sorted by normalized comment score. Similarly, if the user selects the normalized average raw comment score then a list is produced which has grouped the comments sorted by normalized average comment score.

At this point, the user has a preliminary set of results. A user can elect to remove raw words from the list of raw words that have been preselected in the first criteria measurement. If the user chooses to ignore these words then the method begins again with the selected words ignored. Otherwise the method is complete.

An exemplarily embodiment of the invention is shown below. Pepsi MAX posted this question to their Facebook wall on Dec. 18 2010, “Okay—You've got 30 seconds to tell a friend why they should try Pepsi MAX. WHAT DO YOU SAY??” They received 145 comments in total.

The raw words selected were:

The normalized word scores were:

The unique word scores were:

The raw comment scores were:

The normalized comment scores were:

The average raw comment scores were:

The normalized average comment scores were:

The listed comments were:

and: