Emotion-based voice controlled device

1. A computer-implemented method comprising:

receiving, at a voice controlled device, a voice command from a user;

receiving, at a server computer system comprising a processor, memory accessible by the processor, and computer program instructions stored in the memory and executable by the processor, the voice command and converting the voice command to text data representing the voice command;

obtaining, at the server computer system, a plurality of items of content responsive to the voice command by searching for content using the text;

determining, at the server computer system, at least one class related to the voice command;

classifying, at the server computer system, each obtained item of content into at least one class, wherein each obtained item of content is classified into at least one class using text data associated with each item of content by:

counting, at the server computer system, occurrences of words in the text data associated with each item of content, wherein the counted words are included in a dictionary for each class including a ranking of emotional words,

determining, at the server computer system, an apparent number of occurrences of each word in the text data associated with each item of content,

determining, at the server computer system, a contextual weight of each word in the text data associated with each item of content with respect to the item of content,

determining, at the server computer system, a weight of each word in the text data associated with each item of content;

determining, at the server computer system, a score for each class for each item of content based on the counted occurrences of words, the determined apparent number of occurrences of each word, the determined contextual weight of each word, and the determined weight of each word and

classifying, at the server computer system, each item of content into at least one class based on the determined score for each class;

identifying, at the server computer system, at least one item of content classified into at least one class related to the voice command; and

transmitting, at the server computer system, the at least one identified item of content.

2. The method of claim 1 , further comprising converting the speech data to text and the search for content is performed using the text.

3. The method of claim 2 , wherein the classes relate to emotional states and the at least one class related to the voice command is an emotional state of the user determined by analyzing the text.

4. The method of claim 3 , wherein the dictionary is obtained by:

for each class, building, at the server computer system, a dictionary including a minimum set of words that determines the class;

building, at the server computer system, a corpus of text data associated with a plurality of items of content;

searching, at the server computer system, the corpus of text data to find words neighboring or related to word in the dictionary; and

extending, at the server computer system, the dictionary to include the words found in the search.

5. The method of claim 4 , further comprising repeating the building, searching, and extending until all words in the corpus have relative importance scores based on the classes.

6. A server computer system comprising a processor, memory accessible by the processor, and computer program instructions stored in the memory and executable by the processor to perform:

receiving a voice command from a user of the voice controlled device from the voice controlled device at the server computer system and converting the voice command to text data representing the voice command;

obtaining a plurality of items of content responsive to the voice command by searching for content using the text data;

determining at least one class related to the voice command;

classifying each obtained item of content into at least one class, wherein each obtained item of content is classified into at least one class using text data associated with each item of content by:

counting, at the computer system, occurrences of words in the text data associated with each item of content, wherein the counted words are included in a dictionary for each class including a ranking of emotional words,

determining, at the computer system, an apparent number of occurrences of each word in the text data associated with each item of content,

determining, at the computer system, a contextual weight of each word in the text data associated with each item of content with respect to the item of content,

determining, at the computer system, a weight of each word in the text data associated with each item of content;

determining, at the computer system, a score for each class for each item of content based on the counted occurrences of words, the determined apparent number of occurrences of each word, the determined contextual weight of each word, and the determined weight of each word and

classifying, at the computer system, each item of content into at least one class based on the determined score for each class;

identifying at least one item of content classified into at least one class related to the voice command; and

transmitting the at least one identified item of content.

7. The system of claim 6 , further comprising converting the speech data to text and the search for content is performed using the text.

8. The system of claim 7 , wherein the classes relate to emotional states and the at least one class related to the voice command is an emotional state of the user determined by analyzing the text.

9. The system of claim 8 , wherein the dictionary is obtained by:

for each class, building a dictionary including a minimum set of words that determines the class;

building a corpus of text data associated with a plurality of items of content;

searching the corpus of text data to find words neighboring or related to word in the dictionary; and

extending the dictionary to include the words found in the search.

10. The system of claim 9 , further comprising repeating the building, searching, and extending until all words in the corpus have relative importance scores based on the classes.

11. A computer program product comprising a non-transitory computer readable storage having program instructions embodied therewith, the program instructions executable by a server computer system comprising a processor, memory accessible by the processor, and computer program instructions stored in the memory and executable by the processor, to cause the server computer system to perform a method comprising:

receiving, at the computer system, a voice command from a user of the voice controlled device from the voice controlled device at the server computer system and converting the voice command to text data representing the voice command;

obtaining, at the server computer system, a plurality of items of content responsive to the voice command by searching for content using the text data;

determining, at the server computer system, at least one class related to the voice command;

classifying, at the server computer system, each obtained item of content into at least one class, wherein each obtained item of content is classified into at least one class using text data associated with each item of content by:

counting, at the server computer system, occurrences of words in the text data associated with each item of content, wherein the counted words are included in a dictionary for each class including a ranking of emotional words,

determining, at the server computer system, an apparent number of occurrences of each word in the text data associated with each item of content,

determining, at the server computer system, a contextual weight of each word in the text data associated with each item of content with respect to the item of content,

determining, at the server computer system, a weight of each word in the text data associated with each item of content;

determining, at the server computer system, a score for each class for each item of content based on the counted occurrences of words, the determined apparent number of occurrences of each word, the determined contextual weight of each word, and the determined weight of each word and

classifying, at the server computer system, each item of content into at least one class based on the determined score for each class;

identifying, at the server computer system, at least one item of content classified into at least one class related to the voice command; and

transmitting, at the server computer system, the at least one identified item of content.

12. The computer program product of claim 11 , further comprising converting the speech data to text and the search for content is performed using the text.

13. The computer program product of claim 12 , wherein the classes relate to emotional states and the at least one class related to the voice command is an emotional state of the user determined by analyzing the text.

14. The computer program product of claim 13 , wherein the dictionary is obtained by:

for each class, building, at the server computer system, a dictionary including a minimum set of words that determines the class;

building, at the server computer system, a corpus of text data associated with a plurality of items of content;

searching, at the server computer system, the corpus of text data to find words neighboring or related to word in the dictionary;

extending, at the server computer system, the dictionary to include the words found in the search; and

repeating the building, searching, and extending until all words in the corpus have relative importance scores based on the classes.