exam questions

Exam AI-900 All Questions

View all questions & answers for the AI-900 exam

Exam AI-900 topic 1 question 34 discussion

Actual exam question from Microsoft's AI-900
Question #: 34
Topic #: 1
[All AI-900 Questions]

HOTSPOT -
For each of the following statements, select Yes if the statement is true. Otherwise, select No.
NOTE: Each correct selection is worth one point.
Hot Area:

Show Suggested Answer Hide Answer
Suggested Answer:
Box 1: Yes -
In machine learning, if you have labeled data, that means your data is marked up, or annotated, to show the target, which is the answer you want your machine learning model to predict.
In general, data labeling can refer to tasks that include data tagging, annotation, classification, moderation, transcription, or processing.

Box 2: No -

Box 3: No -
Accuracy is simply the proportion of correctly classified instances. It is usually the first metric you look at when evaluating a classifier. However, when the test data is unbalanced (where most of the instances belong to one of the classes), or you are more interested in the performance on either one of the classes, accuracy doesn't really capture the effectiveness of a classifier.
Reference:
https://www.cloudfactory.com/data-labeling-guide
https://docs.microsoft.com/en-us/azure/machine-learning/studio/evaluate-model-performance

Comments

Chosen Answer:
This is a voting comment (?). It is better to Upvote an existing comment if you don't have anything to add.
Switch to a voting comment New
gargaditya
Highly Voted 3 years, 12 months ago
Answer is correct. Note points: --'labels' are the outputs(ie we map each set of features to a 'label') --recall and precision are other imp metrics in a classification problem(other than Accuracy)
upvoted 28 times
...
ealcober
Highly Voted 4 years ago
I think the first would be FEATURING (feature is input, label is output)
upvoted 8 times
xiban
3 years ago
i had the same doubt but i found the term "labelling data" and its not the same than label target...
upvoted 3 times
...
...
saema
Most Recent 3 months, 1 week ago
Yes NO NO
upvoted 1 times
...
b9b193d
7 months, 1 week ago
data labeling (also known as data annotation or tagging) is indeed the process of adding tags or labels to raw data such as images, videos, text, and audio. These labels provide context and meaning to the data, enabling machine learning algorithms to learn and make predictions based on the annotated information.
upvoted 2 times
...
zellck
1 year, 4 months ago
YNN is the answer. https://learn.microsoft.com/en-us/training/modules/create-regression-model-azure-machine-learning-designer/2-regression-scenarios Regression is a form of machine learning used to understand the relationships between variables to predict a desired outcome. Regression predicts a numeric label or outcome based on variables, or features. For example, an automobile sales company might use the characteristics of a car (such as engine size, number of seats, mileage, and so on) to predict its likely selling price. In this case, the characteristics of the car are the features, and the selling price is the label.
upvoted 2 times
zellck
1 year, 4 months ago
https://learn.microsoft.com/en-us/training/modules/create-regression-model-azure-machine-learning-designer/5-regression-steps To train a regression model, you need a dataset that includes historical features, characteristics of the entity for which you want to make a prediction, and known label values. The label is the quantity you want to train a model to predict. It's common practice to train the model using a subset of the data, while holding back some data with which to test the trained model. This enables you to compare the labels that the model predicts with the actual known labels in the original dataset.
upvoted 1 times
...
...
rdemontis
1 year, 4 months ago
correct answer. For the third sentence: Accuracy is a commonly used metric, especially in binary classification problems, where it measures the percentage of correctly classified instances out of the total. However, in many scenarios, accuracy alone may not provide a comprehensive evaluation of the model's performance, particularly when dealing with imbalanced datasets or when different types of errors have varying degrees of importance. Different evaluation metrics may be more appropriate depending on the problem. For example, in medical diagnostics, sensitivity and specificity (true positive rate and true negative rate) may be more relevant metrics to measure the model's ability to correctly identify positives and negatives. In regression problems, metrics like mean squared error (MSE) or mean absolute error (MAE) are often used. Other factors, such as precision, recall, F1 score, area under the receiver operating characteristic curve (AUC-ROC), and others, are commonly used metrics depending on the problem and the desired evaluation criteria. (ChatGPT)
upvoted 3 times
...
master_yoda
1 year, 6 months ago
No, accuracy is not always the primary metric used to measure a model's performance. There are several metrics that can be used to evaluate the quality of a model's predictions, and the choice of metric depends on the specific problem and goals of the model. Some commonly-used metrics for evaluating model performance include loss, accuracy, precision, recall, and area under the ROC curve (AUC) ². The appropriate metric to use depends on the specific problem and goals of the model. Source: Conversation with Bing, 4/24/2023 (1) Evaluate Models Using Metrics | Machine Learning - Google Developers. https://developers.google.com/machine-learning/testing-debugging/metrics/metrics Accessed 4/24/2023. (2) Performance Metrics in Machine Learning [Complete Guide]. https://neptune.ai/blog/performance-metrics-in-machine-learning-complete-guide Accessed 4/24/2023.
upvoted 3 times
...
vhogstad
1 year, 8 months ago
Answer is correct. I will use the following to remember the NO`s in this question: You should = NO Is always = NO
upvoted 2 times
ColoradoChick
1 year, 3 months ago
Until MS changes the wording slightly to something that's a true statement. This question is a very very (I'll add another very) basic concept in AI. If you need to remember answers based on the exact wording of a question, what will you do in a tech interview when they see the cert on your resume, but you can't explain the concept at all? ..... I know this guy won't see my response. My comment is food for thought for anyone else with this viewpoint. Yes ... your Mom sent me. :))
upvoted 5 times
...
...
Eltooth
2 years, 3 months ago
Yes No No
upvoted 3 times
...
ydu7312
2 years, 5 months ago
Answer is correct
upvoted 2 times
...
heenak55
2 years, 6 months ago
In test on 17 April 2022
upvoted 1 times
...
Pravda
2 years, 7 months ago
On exam 3.3.2022
upvoted 2 times
...
jumata32
2 years, 9 months ago
this came on my exam 1/27/2022 pass with 985/1000, answer is correct, only failed one for NLP
upvoted 1 times
...
MohammadKhubeb
2 years, 9 months ago
Recall, precision, F-measure/F1-score, sensitivity, specificity, AUC of ROC, etc, are other key parameters to evaluate. Also, when the data is highly imbalance then it gets fail. For more info on class imbalance please refer to my detailed article: https://link.springer.com/article/10.1007/s12008-020-00715-3
upvoted 1 times
...
DChilds
2 years, 10 months ago
On question 2, does Microsoft consider the split data (testing and evaluation) from the same dataset, separate data? I assumed that since one dataset can be split for different uses, you are still using the same data? So my suggestion is Yes, Yes, No. Thoughts anyone?
upvoted 1 times
ColoradoChick
1 year, 3 months ago
I understand DChilds' thought process but for everything I've seen in the MS AI & ML docs: dataset/data source - before the Split data = after the split: training data or evaluation data so "same data" is referring to data after the Split, and not the source file/DB/etc that you start with
upvoted 1 times
...
TJ001
2 years, 9 months ago
not quite-- it is identical schema but values are difference that is whole purpose of splitting it by rows
upvoted 1 times
...
...
Marski
2 years, 11 months ago
Isn't a first metric a primary metric? Hmm..
upvoted 1 times
ColoradoChick
1 year, 3 months ago
There is a subtle difference: primary metric = the main metric you used to decide whether or not a model performed well (after looking at & analyzing all the metrics) first metric = the 1st metric you looked at (before you looked at any other metrics or did any analysis)
upvoted 1 times
...
...
abelk
2 years, 11 months ago
I just validated and it is correct
upvoted 1 times
...
Community vote distribution
A (35%)
C (25%)
B (20%)
Other
Most Voted
A voting comment increases the vote count for the chosen answer by one.

Upvoting a comment with a selected answer will also increase the vote count towards that answer by one. So if you see a comment that you already agree with, you can upvote it instead of posting a new comment.

SaveCancel
Loading ...
exam
Someone Bought Contributor Access for:
SY0-701
London, 1 minute ago