IDENTIFYING ENTITIES TRENDING IN A PROFESSIONAL COMMUNITY
A system, method, and apparatus are provided for identifying entities trending within a professional community, such as member of a professional social network. The system collects “share” activity and/or other types of activities conducted by members of the community in which they generate or disseminate (textual) content. From the collected share activity, trending terms are identified and ranked according to scores that reflect the change in frequency of usage of the terms over time. The most relevant shares for each trending term are identified and used to identify names of entities that correspond to (e.g., include) the terms. Reasons indicating why each trending entity is trending are also derived from the share activity. A display or presentation is provided of top trending entities, within one or more segments of the professional community, which includes the reasons and allows a viewer to quickly identify the reason a given entity is trending.
1 . A method, comprising:
recording textual content disseminated by members of a professional community;
identifying trending terms within the recorded textual content;
for each trending term:
identifying content disseminations that are most relevant to the trending term; and
within the relevant content disseminations, identifying at least one trending entity having a name that includes the trending term; and
presenting to a first member a display comprising:
multiple trending entities; and
for each trending entity, at least one reason the trending entity is trending.
2 . The method of claim 1 , wherein said identifying trending terms comprises:
collecting words used in the recorded content disseminations;
for each of multiple collected words, calculating a ratio of a short-term usage of the word to a long-term usage of the word; and
identifying N words having the highest calculated ratios (N>1);
wherein the trending terms are the N words.
3 . The method of claim 2 , wherein each of the short-term usage and the long-term usage is augmented by a smoothing factor prior to calculating the ratio.
4 . The method of claim 2 , wherein:
the short-term is approximately 24 hours; and
the long-term is approximately 7 days.
5 . The method of claim 1 , wherein identifying content disseminations relevant to a given trending term comprises:
collecting words used in content disseminations that mention the given trending term; and
for each content dissemination that mentions the given trending term, calculating a similarity between the content dissemination and the collected words;
wherein the M content disseminations (M>1) having the highest similarity to the collected words are the most relevant content disseminations.
6 . The method of claim 1 , further comprising, for each of multiple trending entities:
selecting, as the reason the trending entity is trending, text of a content dissemination that mentions the trending entity.
7 . The method of claim 1 , wherein the content disseminations comprise share activities of members of a social network.
8 . The method of claim 1 , wherein:
said recording content disseminations comprises, for each of multiple segments of the professional community, separately collecting content disseminations initiated by members of the multiple segments; and
said presenting comprises separately presenting top trending entities of one or more segments specified by the first member.
9 . The method of claim 8 , wherein the multiple segments are associated with different values of one or more of the following member attributes:
industry;
seniority;
function; and
organization size.
10 . An apparatus, comprising:
one or more processors; and
memory storing instructions that, when executed by the one or more processors, cause the apparatus to:
record textual content disseminated by members of a professional community;
identify trending terms within the recorded textual content;
for each trending term:
identify content disseminations that are most relevant to the trending term; and
within the relevant content disseminations, identify at least one trending entity having a name that includes the trending term; and
present to a first member a display comprising:
multiple trending entities; and
for each trending entity, at least one reason the trending entity is trending.
11 . The apparatus of claim 10 , wherein said identifying trending terms comprises:
collecting words used in the recorded content disseminations;
for each of multiple collected words, calculating a ratio of a short-term usage of the word to a long-term usage of the word; and
identifying N words having the highest calculated ratios (N>1);
wherein the trending terms are the N words.
12 . The apparatus of claim 11 , wherein each of the short-term usage and the long-term usage is augmented by a smoothing factor prior to calculating the ratio.
13 . The apparatus of claim 11 , wherein:
the short-term is approximately 24 hours; and
the long-term is approximately 7 days.
14 . The apparatus of claim 10 , wherein identifying content disseminations relevant to a given trending term comprises:
collecting words used in content disseminations that mention the given trending term; and
for each content dissemination that mentions the given trending term, calculating a similarity between the content dissemination and the collected words;
wherein the M content disseminations (M>1) having the highest similarity to the collected words are the most relevant content disseminations.
15 . The apparatus of claim 10 , wherein the memory further stores instructions that, when executed by the one or more processors, cause the apparatus to, for each of multiple trending entities:
select, as the reason the trending entity is trending, text of a content dissemination that mentions the trending entity.
16 . The apparatus of claim 10 , wherein the content disseminations comprise share activities of members of a social network.
17 . The apparatus of claim 10 , wherein:
said recording content disseminations comprises, for each of multiple segments of the professional community, separately collecting content disseminations initiated by members of the multiple segments; and
said presenting comprises separately presenting top trending entities of one or more segments specified by the first member.
18 . The apparatus of claim 17 , wherein the multiple segments are associated with different values of one or more of the following member attributes:
industry;
seniority;
function; and
organization size.
19 . A system, comprising:
one or more processors; and
a trend identification module comprising a non-transitory computer-readable medium storing instructions that, when executed, cause the system to:
record textual content disseminated by members of a professional community;
identify trending terms within the recorded textual content;
for each trending term:
identify content disseminations that are most relevant to the trending term; and
within the relevant content disseminations, identify at least one trending entity having a name that includes the trending term; and
present to a first member a display comprising:
multiple trending entities; and
for each trending entity, at least one reason the trending entity is trending.
20 . The apparatus of claim 10 , wherein the non-transitory computer-readable medium further stores instructions that, when executed by the one or more processors, cause the system to, for each of multiple trending entities:
select, as the reason the trending entity is trending, text of a content dissemination that mentions the trending entity.