A Text Data Analysis Method Involving Emotion in Computer
A technology of text data and analysis methods, applied in text database clustering/classification, unstructured text data retrieval, calculation, etc., can solve problems such as inaccurate output results and difficulty in determining the number of topics
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0107] In this embodiment, for the online comment corpus of the laptop computer, a corresponding artificial target language is constructed to form a parallel corpus for training the alignment model.
[0108] For a comment sentence: "Fantastic for the price, it's a pity keys were notilluminated." (rough meaning: the price is right, unfortunately the keyboard is not backlit.) The marked results include two Opinions: , . Construct artificial language sentences according to the grammar explained above: "LAPTOP PRICE is positive; KEYBOARD DESIGN_FEATURES is negative;".
Embodiment 2
[0110] In this embodiment, the system accepts a natural language sentence and outputs two-tuple information. The process is as follows:
[0111] 1. The system receives a natural language sentence: The Dell is quick enough but not good with finger prints.
[0112] 2. The system extracts the word information vector Words;
[0113] 3. The system calculates the alignment probability feature vector AlignmentProbabilities according to the alignment probability table;
[0114] 4. The system extracts phrase feature vector Phrases according to the extracted phrase table;
[0115] 5. Input the features of steps 2, 3, and 4 into the aspect recognition model, and find two aspect categories whose output scores are higher than the threshold of 0.14: LAPTOP#OPERATION_PERFORMANCE, LAPTOP#QUALITY;
[0116] 6. The system judges that the emotional tendency of LAPTOP#OPERATION_PERFORMANCE in the sentence is positive;
[0117] 7. The system judges that the emotional tendency of LAPTOP#QUALITY ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


