A Data Scientist Has Developed a Machine Learning Translation Model

question 153

Multiple Choice

A data scientist has developed a machine learning translation model for English to Japanese by using Amazon SageMaker's built-in seq2seq algorithm with 500,000 aligned sentence pairs. While testing with sample sentences, the data scientist finds that the translation quality is reasonable for an example as short as five words. However, the quality becomes unacceptable if the sentence is 100 words long. Which action will resolve the problem?

• A) Change preprocessing to use n-grams.
• B) Add more nodes to the recurrent neural network (RNN) than the largest sentence's word count.
• C) Adjust hyperparameters related to the attention mechanism.
• D) Choose a different weight initialization type.

---

Definitions:

Intermodal Matching

The process of coordinating and integrating different modes of transportation for efficient cargo or passenger transfer.

Facial Expressions

Visible manifestations of emotions and feelings through facial movements, serving as important non-verbal communication cues.

Voice and Face

Fundamental aspects of human identity and social interaction, where voice refers to vocal characteristics and communications, and face refers to facial expressions and recognition.

Most Developed

Refers to countries or regions that have achieved the highest levels of industrialization, technological advancement, and standard of living.