BANL 1100 Review [English]

What is the first step in analyzing a dataset like the Baidu search logs?

• Create visualizations

• Run statistical tests

• Clean and prepare the data

• Write the final report

Answer explanation

The first step in analyzing a dataset like the Baidu search logs is to clean and prepare the data. This ensures that the data is accurate and usable for further analysis, such as creating visualizations or running statistical tests.

Which of the following is NOT typically part of the data cleaning process?

• Handling missing values

• Converting data types

• Creating a data dictionary

• Performing regression analysis

Answer explanation

Performing regression analysis is a statistical method used for prediction, not a data cleaning task. Data cleaning involves handling missing values, converting data types, and creating a data dictionary.

How do I create a histogram in R to visualize the distribution of response times across different device types?

• Use the plot() function

• Use the hist() function with facet_wrap()

• Use the barplot() function

• Use the boxplot() function

Answer explanation

To visualize the distribution of response times across different device types, use the hist() function with facet_wrap() to create separate histograms for each device type, allowing for clear comparison.

What statistical test should I use to compare click-through rates between mobile and desktop users?

• T-test

• ANOVA

• Chi-square test

• Z-test

Answer explanation

A Z-test is appropriate for comparing click-through rates between two independent groups (mobile and desktop users) when the sample size is large and the data is normally distributed. It assesses the difference in proportions effectively.

How can I use ggplot2 to create a scatter plot showing the relationship between query complexity and time spent on search result pages?

• Use geom_bar()

• Use geom_line()

• Use geom_point()

• Use geom_boxplot()

Answer explanation

To visualize the relationship between query complexity and time spent, use geom_point() in ggplot2. This function creates a scatter plot, effectively displaying how these two variables correlate.

Which R packages are best for cleaning and preprocessing the Baidu search log data?

• ggplot2 and lattice

• dplyr and tidyr

• caret and glmnet

• rpart and randomForest

Answer explanation

dplyr and tidyr are ideal for cleaning and preprocessing data due to their powerful functions for data manipulation and reshaping. ggplot2 and lattice are for visualization, while caret, glmnet, rpart, and randomForest focus on modeling.

What's the most appropriate way to visualize trends in search volume over time using R?

• Histogram

• Scatter plot

• Box plot

• Line plot

Answer explanation

A line plot is the most appropriate way to visualize trends in search volume over time, as it effectively shows changes and patterns across a continuous time scale, making it easy to interpret trends.

How do I perform a chi-square test in R to analyze the relationship between user demographics and query types?

• Use the t.test() function

• Use the aov() function

• Use the chisq.test() function

• Use the lm() function

Answer explanation

To analyze the relationship between user demographics and query types, the correct function to use in R is chisq.test(). This function performs the chi-square test, which is suitable for categorical data analysis.

What online resources can I use to research current trends in Chinese search engine user behavior?

• Google Trends

• Baidu Index

• Bing Webmaster Tools

• Yandex Metrica

Answer explanation

Baidu Index is the most relevant resource for researching trends in Chinese search engine user behavior, as it specifically tracks data from Baidu, China's leading search engine. Other options focus on different markets.

Which function in R is suitable for visualizing the distribution of a dataset across different categories?

• Use pie() function

• Use scatterplot() function

• Use boxplot() function

• Use lineplot() function

Answer explanation

The boxplot() function is ideal for visualizing the distribution of a dataset across different categories, as it displays the median, quartiles, and potential outliers, making it easier to compare distributions.