Searching for ideograms in an online social network

Patent metadata

Field	Value
Publication number	US-9721024-B2
Application number	US-201414576926-A
Country	US
Kind code	B2
Filing date	Dec 19, 2014
Priority date	Dec 19, 2014
Publication date	Aug 1, 2017
Grant date	Aug 1, 2017

Abstract

In one embodiment, a method includes receiving a query inputted at a mobile device of a user of an online social network. The query may include n-grams (e.g., words, emoticons, and punctuation marks), and may be, for example, a message to another user or search terms. N-grams may be mapped to tags, which may, in turn, be mapped to ideograms (i.e., stickers). One or more ideograms may be identified based on the one or more ideogram being associated with tags matching n-grams of the query. For each of the identified ideograms, a probability quantifying how likely the user is to select the ideogram is calculated based on how frequently the ideogram has been used. A set of ideogram suggestions, including one or more of the identified ideograms, is generated based on the probabilities and sent to the user's mobile device for display to the user.

First claim

What is claimed is:

1. A method comprising, by one or more computing devices:
receiving a first query from a first client system of a first user of an online social network, wherein the first query is inputted by the first user at the first client system, the first query comprising one or more n-grams, and wherein the online social network comprises a social graph comprising a plurality of nodes and a plurality of edges connecting the nodes, each of the edges between two of the nodes representing a single degree of separation between two of the nodes, plurality of nodes comprising:
a first node corresponding to the first user of the online social network; and
a plurality of second nodes that each correspond to a second user or a concept associated with the online social network;

identifying one or more ideograms, each ideogram being associated with one or more tags, each identified ideogram being associated with at least one tag matching at least one of the n-grams of the received first query;
calculating, for each identified ideogram, a use-probability for the ideogram given the received first query, wherein the use-probability is based at least in part on a frequency of use associated with the ideogram; and
sending, to the first client system, a first set of ideograms comprising one or more of the identified ideograms for display to the first user, the first set being determined based on the calculated use-probabilities associated with the ideograms.

2. The method of claim 1 , wherein at least one of the n-grams corresponds to an emoticon.

3. The method of claim 1 , wherein the frequency of use is determined for a plurality of users of the online social network.

4. The method of claim 1 , wherein the frequency of use is determined for the first user.

5. The method of claim 1 , wherein s denotes an ideogram, denotes the first query, and the use-probability, P(s|q) , is computed as:

P
⁡

(

s
❘
q

)

=

P
⁡

(

q
❘
s

)

·

P
⁡

(
s
)

P
⁡

(
q
)

=

P
⁡

(

q
❘
s

)

·

P
⁡

(
s
)

⁢

⁢
wherein

[
1
]

P
⁡

(

q
❘
s

)

=

number of times
s
was sent for
q

number of times
s
was sent for any query

,
and

[
2
]

P
⁡

(
s
)

=

number of times
s
was sent

number of times any sticker was sent

.

[
3
]

6. The method of claim 1 , further comprising determining one or more tags that match at least one of the n-grams of the received first query based on natural-language processing of the received first query.

7. The method of claim 1 , further comprising:
receiving, before receipt of the first query, a second query from a second client system of a second user of the online social network, the second query inputted by the second user at the second client system, the second query comprising one or more n-grams; and
identifying one or more ideograms, each identified ideogram associated with at least one tag matching at least one of the n-grams of the received second query.

8. The method of claim 7 , wherein the second user is connected to the first user within the social graph by a threshold degree of separation.

9. The method of claim 1 , further comprising:
ranking each of the identified ideograms based on the

Assignees

• Facebook Inc

Inventors

• Sa Brian Paihan
• Davis Stephen Moore
• Vilimonovic Nikola
• Tan Jireh Yiwei
• Goldsmid Aaron Payne

Classifications

• G06F16/9535 Primary

  Search customisation based on user profiles and personalisation · CPC title

• G06F16/9536 Primary

  Search customisation based on social or collaborative filtering · CPC title

• G06F17/30867 Primary

  Physics · mapped topic

• G06F17/30687

  Physics · mapped topic

• G06F16/3346

  using probabilistic model · CPC title