System and method for assessing and correcting potential underserved content in natural language understanding applications
Methods, systems, and related products that provide detection of media content items that are under-locatable by machine voice-driven retrieval of uttered requests for retrieval of the media items. For a given media item, a resolvability value and/or an utterance resolve frequency is calculated by a number of playbacks of the media item by a speech retrieval modality to a total number of playbacks of the media item regardless of retrieval modality. In some examples, the methods, systems and related products also provide for improvement in the locatability of an under-locatable media item by collecting and/or generating one or more pronunciation aliases for the under-locatable item.
1. A method of detecting an under-locatable media content item, comprising:
retrieving, from a database, a name entity associated with a media content item, the name entity having a name entity text; and
determining, based on a name entity classification tag associated with the name entity, that the name entity is under-locatable by a machine voice-driven retrieval of a playback command utterance commanding playback of the media content item, the name entity classification tag being based on the name entity text; and
when the name entity is determined to be under-locatable:
forming a task item based on the name entity, and
passing the task item to an alias generation engine to cause the alias generation engine to classify the media content item.
2. The method of claim 1 , further comprising:
tagging the name entity with the name entity classification tag.
3. The method of claim 2 ,
wherein the tagging is based on a crowd-sourced pronunciation alias for the name entity.
4. The method of claim 2 , wherein the tagging is based on the name entity text including one of:
a neologism;
an abbreviation;
an acronym;
a number;
a date;
a time;
a removal of a space;
a vocable;
a non-replacement symbol;
an orthographically similar replacement symbol;
a semantically similar replacement symbol;
an expressive spelling;
an alternative spelling;
a homophone;
a pun;
a portmanteau; and
a proper noun.
5. The method of claim 1 , further comprising:
retrieving, from the database, another name entity associated with another media content item, wherein data generated from playbacks of the another media content item indicate that the another media content item is under-locatable; and
determining that the another name entity is not tagged with a name entity classification tag and, based thereon, determining that the another media content item is locatable.
6. The method of claim 5 ,
wherein the data is based on a number of playbacks of the media content item triggered by a machine voice-driven retrieval of a playback command utterance.
7. A system for detecting an under-locatable media content item, comprising:
one or more processors adapted to:
retrieve, from a database, a name entity associated with a media content item, the name entity having a name entity text; and
determine, based on a name entity classification tag associated with the name entity, that the name entity is under-locatable by a machine voice-driven retrieval of a playback command utterance commanding playback of the media content item, the name entity classification tag being based on the name entity text; and
when the name entity is determined to be under-locatable:
form a task item based on the name entity, and
pass the task item to an alias generation engine to cause the alias generation engine to classify the media content item.
8. The system of claim 7 , the one or more processors further adapted to:
tag the name entity with the name entity classification tag.
9. The system of claim 8 ,
wherein the tagging is based on a crowd-sourced pronunciation alias for the name entity.
10. The system of claim 8 , wherein the tagging is based on the name entity text including one of:
a neologism;
an abbreviation;
an acronym;
a number;
a date;
a time;
a removal of a space;
a vocable;
a non-replacement symbol;
an orthographically similar replacement symbol;
a semantically similar replacement symbol;
an expressive spelling;
an alternative spelling;
a homophone;
a pun;
a portmanteau; and
a proper noun.
11. The system of claim 7 , the one or more processors further operable to:
retrieve, from the database, another name entity associated with another media content item, wherein data generated from playbacks of the another media content item indicate that the another media content item is under-locatable; and
determine that the another name entity is not tagged with a name entity classification tag and, based thereon, determining that the another media content item is locatable.
12. The system of claim 11 ,
wherein the data is based on a number of playbacks of the media content item triggered by a machine voice-driven retrieval of a playback command utterance.
13. A non-transitory computer-readable medium having stored thereon sequences of instructions, the sequences of instructions including instructions which when executed by a computer system causes the computer system to perform:
retrieving, from a database, a name entity associated with a media content item, the name entity having a name entity text; and
determining, based on a name entity classification tag associated with the name entity, that the name entity is under-locatable by a machine voice-driven retrieval of a playback command utterance commanding playback of the media content item, the name entity classification tag being based on the name entity text; and
when the name entity is determined to be under-locatable:
forming a task item based on the name entity, and
passing the task item to an alias generation engine to cause the alias generation engine to classify the media content item.
14. The non-transitory computer-readable medium according to claim 13 , further comprising:
tagging the name entity with the name entity classification tag.
15. The non-transitory computer-readable medium according to claim 14 ,
wherein the tagging is based on a crowd-sourced pronunciation alias for the name entity.
16. The non-transitory computer-readable medium according to claim 14 , wherein the tagging is based on the name entity text including one of:
a neologism;
an abbreviation;
an acronym;
a number;
a date;
a time;
a removal of a space;
a vocable;
a non-replacement symbol;
an orthographically similar replacement symbol;
a semantically similar replacement symbol;
an expressive spelling;
an alternative spelling;
a homophone;
a pun;
a portmanteau; and
a proper noun.
17. The non-transitory computer-readable medium according to claim 13 , further comprising:
retrieving, from the database, another name entity associated with another media content item, wherein data generated from playbacks of the another media content item indicate that the another media content item is under-locatable; and
determining that the another name entity is not tagged with a name entity classification tag and, based thereon, determining that the another media content item is locatable.
18. The non-transitory computer-readable medium according to claim 17 ,
wherein the data is based on a number of playbacks of the media content item triggered by a machine voice-driven retrieval of a playback command utterance.