Electronic Device Processing User Utterance for Image Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Intelligent services struggle to provide emotional satisfaction to users by only fulfilling the requested search conditions, lacking the ability to offer additional information that could enhance user experience.
Innovation Solution
An electronic device and method that processes user utterances to classify images and provide additional attribute information, allowing for emotional engagement by offering contextually relevant additional information beyond the initial search request, such as emotional states and visual effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the intelligent application only provides information matching the user's search conditions, then the search accuracy is improved, but the user's emotional satisfaction deteriorates
Solution Approach 1:
The patent segments the information provision into two distinct parts: (1) precise search results matching user conditions, and (2) additional emotional context information. This segmentation allows the system to maintain search accuracy while separately delivering emotional satisfaction through supplementary content like emotional states and situational context.
Solution Approach 2:
The patent adds a new dimension to the information provision by introducing emotional attributes and contextual information beyond the basic search parameters. This transforms the service from purely functional (matching search conditions) to include emotional and contextual dimensions, thereby satisfying user emotional needs without compromising search precision.
2Adaptability or versatility
If the electronic device provides additional attribute information beyond search conditions, then the user's emotional satisfaction is improved, but the information processing complexity increases
Solution Approach 1:
The system performs preliminary actions by pre-processing images to extract emotional attributes and contextual information before the user even makes a search request. This advance preparation stores emotional metadata that can be quickly retrieved and paired with search results, reducing the processing burden during actual user interactions.
Solution Approach 2:
The patent introduces an intermediary layer of emotional attribute extraction and context analysis that sits between the basic image database and the user interface. This intermediary processes images to generate emotional metadata, which then enriches the search results without requiring the entire system to become more complex. The intermediary handles the complexity locally while presenting simplified enriched results to users.
3Adaptability or versatility
If the system classifies images by multiple attributes including emotional states, then the service versatility is improved, but the classification complexity increases
Solution Approach 1:
The classification system is segmented into independent attribute extractors that each handle specific aspects (emotional state, situation, object, etc.). Each extractor operates independently and produces separate classification results that are then combined, allowing the system to achieve multi-attribute classification versatility while managing complexity through modular, independent processing units.
Data Source
AI summary
A user terminal processing a user utterance and a control method thereof are provided. A user terminal according to various embodiments of the disclosure includes a processor configured as a portion of the user terminal or configured to remotely communicate with the user terminal; and a memory configured to operatively connect to the processor, wherein the memory may be configured to store instructions configured, when executed, to enable the processor to, receive a user utterance, the user utterance including a first expression for classifying a plurality of images, transmit information about the received user utterance to an external electronic device using a communication circuit, and perform a task according to operation information by receiving the operation information associated with the user utterance, and the operation information may include an operation of providing the first expression and the second expression indicating attribute information about images classified by the first expression.