You work for a large technology company that wants to modernize their contact center.
You have been asked to develop a solution to classify incoming calls by product so that requests can be more quickly routed to the correct support team.
You have already transcribed the calls using the Speech-to-Text API.
You want to minimize data preprocessing and development time.
How should you build the model?
Click on the arrows to vote for the correct answer
A. B. C. D.A.
The task is to develop a solution to classify incoming calls by product so that requests can be more quickly routed to the correct support team. The calls have already been transcribed using the Speech-to-Text API, and the goal is to minimize data preprocessing and development time. Given this context, there are several possible ways to build the model.
Option A: Use the AI Platform Training built-in algorithms to create a custom model. This option involves using the built-in algorithms provided by AI Platform Training to create a custom model. This approach can be effective if the built-in algorithms are a good fit for the specific use case, as it can reduce the amount of time and effort required to develop the model. However, it is important to ensure that the built-in algorithms are suitable for the specific task of classifying incoming calls by product. Additionally, it may still require some data preprocessing to prepare the data for the model.
Option B: Use AutoML Natural Language to extract custom entities for classification. This option involves using AutoML Natural Language to extract custom entities for classification. AutoML Natural Language is a machine learning service that enables developers to train custom machine learning models for text classification, sentiment analysis, and other natural language processing tasks. This approach can be effective if the data requires custom entity extraction, which may be the case if there are specific product names or terms that need to be identified in the transcribed calls. However, this approach may require more development time than some of the other options.
Option C: Use the Cloud Natural Language API to extract custom entities for classification. This option involves using the Cloud Natural Language API to extract custom entities for classification. The Cloud Natural Language API is a pre-trained machine learning model that provides natural language understanding for applications. This approach can be effective if the data requires custom entity extraction, similar to Option B. However, like Option B, this approach may require more development time than some of the other options.
Option D: Build a custom model to identify the product keywords from the transcribed calls, and then run the keywords through a classification algorithm. This option involves building a custom model to identify the product keywords from the transcribed calls, and then running the keywords through a classification algorithm. This approach can be effective if the data requires custom preprocessing to identify the product keywords, which may be the case if there are specific terms or phrases that need to be identified in the transcribed calls. However, this approach may require more development time than some of the other options, as it involves building a custom model.
In summary, each of the options has its own advantages and disadvantages. The best approach depends on the specific requirements of the project, such as the complexity of the classification task and the amount of development time available. Options A, B, and C can be effective if the data requires custom entity extraction, while Option D may be better suited if the data requires custom preprocessing. Ultimately, the choice of approach should be based on careful consideration of the project requirements and the available resources.