Crowdsourced Translation Priority System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Social networking systems face challenges in crowdsourcing reliable translations due to unreliable translators, including malicious users, well-meaning but incorrect submissions, and the limited number of users and time available for translation tasks, which affects the accuracy and efficiency of translating text strings across languages.
Innovation Solution
The system selects text strings for translation based on priority values determined by previously calculated reliability values, uses credibility scores of users to evaluate and prioritize translation submissions, and introduces a method to spot-check user translations against known correct translations to ensure reliability, while also considering the impact and importance of the text strings and introducing randomness in the selection process to distribute user effort effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the system crowdsources translations from many users, then translation coverage and speed increase, but translation reliability decreases due to malicious users and incorrect submissions
Solution Approach 1:
The system implements feedback mechanisms where users can vote for or against translations, and the system monitors translation quality over time. This feedback loop allows the system to identify reliable translators and adjust priorities accordingly, resolving the contradiction by maintaining high productivity while improving reliability through continuous quality monitoring and user interaction.
Solution Approach 2:
The system changes the parameter of translation selection from random or simple queue-based to priority-based selection, where translations are ordered by reliability values and other factors. This parameter change allows the system to maintain high translation volume while focusing resources on the most reliable translations first, thus improving overall translation accuracy without sacrificing speed.
2Reliability
If the system prioritizes text strings with low reliability values, then translation accuracy improves, but the complexity of evaluating and managing translation priorities increases
Solution Approach 1:
The system performs preliminary actions by pre-calculating and storing reliability values for text strings before translation occurs. This preliminary evaluation allows the system to sort and prioritize translations in advance, reducing the complexity of real-time decision-making while maintaining high accuracy focus.
Solution Approach 2:
The system segments the translation management process into separate evaluation components: reliability value calculation, priority assignment, and translation execution. This segmentation allows each component to be optimized independently, reducing overall system complexity while maintaining the ability to prioritize accurate translations.
3Reliability
If the system uses credibility scores to evaluate user translations, then translation reliability improves, but the computational overhead and time required for evaluation increases
Solution Approach 1:
The system calculates and stores credibility scores for users in advance, rather than computing them in real-time during translation evaluation. This preliminary action significantly reduces the time required during translation processing while maintaining the ability to filter and prioritize translations based on user trustworthiness.
Solution Approach 2:
The system uses self-service mechanisms where users naturally provide the data needed for credibility assessment through their translation submissions and feedback interactions. This eliminates the need for complex external verification systems while still enabling reliable evaluation of translation quality.
4Adaptability or versatility
If the system introduces randomness in text string selection, then user effort distribution improves, but the predictability and optimization of translation output decreases
Solution Approach 1:
The system implements dynamic selection that combines priority-based ordering with controlled randomness. Translations are primarily selected based on reliability and other static factors, but randomness is introduced as a dynamic element to distribute user effort and prevent systematic biases. This dynamic approach maintains both consistency in quality focus and versatility in user engagement.
Data Source
AI summary
In one embodiment, a method includes accessing, by one or more of the computing devices, one or more translations for each text string of a plurality of text strings; determining, by one or more of the computing devices, a priority value for each text string of the plurality of text strings, wherein the priority value for the text string is based on one or more reliability-values of the one or more translations for the text string; selecting, by one or more of the computing devices, a particular text string from the plurality of text strings based on its priority value; and sending, to a client system, instructions configured to present a translation prompt comprising the particular text string.