Olive Garden vs Buca Di Beppo Reviews

Author

Erin McLaughlin

Olive Garden vs Buca Di Beppo

I want to do a text analysis on the reviews from Olive Garden, located in Oakley, OH and Buca Di Beppo located in Rookwood, OH. This is comparing two Italian Restaurant Chains. Olive Garden is the more well known chain, however Buca is a family style Italian restaurant.

What We Want to Know

We want to see how customers feel about the Olive Garden compared to the Buca Di Beppo in Cincinnati, OH. We want to see if there are any themes or is there an overwhelming positive or negative sentiment.

What Data to Collect?

I am going to collect the 149 most recent Yelp reviews for each restaurant. I am going to evaluate the sentiment of the content in the reviews. The analysis will provide the evidence needed to see the overall “feel” of the two restaurants. I want to filter our contents by looking for words that have positive or negative sentiment.

Plan to Collect the Data

I am collecting the data through using text elements on the Yelp website. I am able to identify what elements I want in my data set through using the html text elements. I then intend to use this data where each row is a review and each review has a name, date, rating, location, and content. (as well as the page ID that references the web page the review is from).

Comparing Sentiments

I then want to look at what words occur most in the reviews associated with Olive Garden and Buca. Therefore I made a word web that looks at the relationships of the words. These webs look at paired words. These webs show how the words are related to each other and how often that pair is used. If a word pair is counted many times in the reviews the line connecting them will increase in size and pigmentation. The words shown are ones that were connected at least 20 times.

After coming the paired words of the most common pairs were food service for both restaurants, and this should be anticipated due to the restaurants being in the food service industry. Time was mentioned on both, but the word wait was mentioned around 20 times for Olive Garden. Further investigation would have to be done to determine if wait times are an issue for Olive Garden and could relate to negative reviews. After looking into the pairing of words, I want to look at whether more positive or negative words are used in the reviews.

The above shows that Buca consistently had positive words mentioned in their reviews. Olive Garden had many negative words included in their reviews. Recommend was mentioned many times by both however, we cannot assume that the 16 reviews, for olive garden, that say recommend were using the word in a negative or positive way. The last way to get an overall feel of the two restaurants is to categorize the words used in the reviews into characteristics.

The above shows that Olive Garden could be reviewed with sentiments of trust and joy. However they also have more sentiments of anger, disgust, fear, and sadness. Buca has more sentiments of anticipation and surprise in their reviews.

Concluding Thoughts

Looking at the sentiment analysis of Olive Garden and Buca Di Beppo there is positive sentiment surrounding the reviews for Buca and a negative sentiment for Olive Garden. Olive Gardens’ reviews had word pairs that included wait, and they have many negative words included in their reviews. Olive Garden also was considered horrible and had significantly more negative reviews than Buca.

--- title: "Olive Garden vs Buca Di Beppo Reviews" author: "Erin McLaughlin" editor: visual toc: true # Generates an automatic table of contents. format: # Options related to formatting. html: # Options related to HTML output. code-tools: TRUE # Allow the code tools option showing in the output. embed-resources: TRUE # Embeds all components into a single HTML file. execute: # Options related to the execution of code chunks. warning: FALSE # FALSE: Code chunk sarnings are hidden by default. message: FALSE # FALSE: Code chunk messages are hidden by default. echo: FALSE # TRUE: Show all code in the output. --- ## Olive Garden vs Buca Di Beppo I want to do a text analysis on the reviews from Olive Garden, located in Oakley, OH and Buca Di Beppo located in Rookwood, OH. This is comparing two Italian Restaurant Chains. Olive Garden is the more well known chain, however Buca is a family style Italian restaurant. ## What We Want to Know We want to see how customers feel about the Olive Garden compared to the Buca Di Beppo in Cincinnati, OH. We want to see if there are any themes or is there an overwhelming positive or negative sentiment. ```{r} #| label: Setting Up Your Data #| include: FALSE # "Run the code, but do not show any code or output thereof" library(tidyverse) # All the tidy things library(lubridate) # Easily fixing pesky dates library(tidytext) # Tidy text mining library(textdata) # Lexicons of sentiment data library(widyr) # Easily calculating pairwise counts library(igraph) # Special graphs for network analysis library(ggraph) reviews<-read_csv("https://www.dropbox.com/scl/fi/pyf76pr6jtp1i0qqe4v6x/erin_italian_restaurants.csv?rlkey=3ygt0ndiyznhtgnsqnf7xeo61&dl=1") ``` ## What Data to Collect? I am going to collect the 149 most recent Yelp reviews for each restaurant. I am going to evaluate the sentiment of the content in the reviews. The analysis will provide the evidence needed to see the overall "feel" of the two restaurants. I want to filter our contents by looking for words that have positive or negative sentiment. ```{r} #| label: Taking Data #| include: FALSE # "Run the code, but do not show any code or output thereof" #the below is our dataframe with the 60 total yelp reviews # to start lets look at the average rating for each restaurant reviews$date<-mdy(reviews$review_date) reviews <- reviews %>% arrange(date) %>% mutate(review_id = row_number()) reviews %>% group_by(restaurant) %>% summarise(n = mean(review_rating)) #Give the Reviews an ID reviews <- reviews %>% arrange(date) %>% mutate(review_id = row_number()) ``` ## Plan to Collect the Data I am collecting the data through using text elements on the Yelp website. I am able to identify what elements I want in my data set through using the html text elements. I then intend to use this data where each row is a review and each review has a name, date, rating, location, and content. (as well as the page ID that references the web page the review is from). ```{r} #| label: Tokeining Data #| include: FALSE tidy_reviews <- reviews %>% unnest_tokens(word, review_content) %>% anti_join(stop_words) ``` ```{r} #| label: Tokeining Data Continued #| include: FALSE review_word_pairs <- tidy_reviews %>% group_by(restaurant) %>% pairwise_count(item = word, # The token vector to count pairs of feature = review_id,# The document vector within which to count upper = FALSE) %>% # Include duplicate pairwise combinations arrange(-n) ``` ## Comparing Sentiments I then want to look at what words occur most in the reviews associated with Olive Garden and Buca. Therefore I made a word web that looks at the relationships of the words. These webs look at paired words. These webs show how the words are related to each other and how often that pair is used. If a word pair is counted many times in the reviews the line connecting them will increase in size and pigmentation. The words shown are ones that were connected at least 20 times. ```{r} #| label: Visualizing words #| include: TRUE bing <- get_sentiments("bing") set.seed(1234) review_word_pairs %>% filter(restaurant == "olive_garden", !item1 == "olive", !item2 == "olive", !item1 == "garden", !item2 == "garden",) %>% ungroup() %>% select(!restaurant) %>% filter(n>=20) %>% graph_from_data_frame() %>% ggraph(layout = "fr") + # "fr" is a type of network graph geom_edge_link(aes(edge_alpha = n, edge_width = n), edge_colour = "#9FA73A") + geom_node_point(size = 3) + geom_node_text(aes(label = name), repel = TRUE, point.padding = unit(0.2, "lines")) + theme_void() ``` ```{r} #| label: Buca Words #| include: TRUE set.seed(1234) review_word_pairs %>% filter(restaurant == "buca") %>% ungroup() %>% select(!restaurant) %>% filter(n>=20) %>% graph_from_data_frame() %>% ggraph(layout = "fr") + # "fr" is a type of network graph geom_edge_link(aes(edge_alpha = n, edge_width = n), edge_colour = "#CB2147") + geom_node_point(size = 3) + geom_node_text(aes(label = name), repel = TRUE, point.padding = unit(0.2, "lines")) + theme_void() ``` After coming the paired words of the most common pairs were food service for both restaurants, and this should be anticipated due to the restaurants being in the food service industry. Time was mentioned on both, but the word wait was mentioned around 20 times for Olive Garden. Further investigation would have to be done to determine if wait times are an issue for Olive Garden and could relate to negative reviews. After looking into the pairing of words, I want to look at whether more positive or negative words are used in the reviews. ```{r} #| label: Bing Comparison #| include: TRUE review_counts <- tidy_reviews %>% group_by(restaurant, word) %>% summarize(n = n()) %>% inner_join(bing) review_counts %>% group_by(restaurant) %>% filter(n>10) %>% mutate(n = ifelse(sentiment == "negative", -n, n)) %>% mutate(word = reorder(word, n)) %>% ggplot(aes(word, n)) + geom_col(fill = "cyan") + coord_flip() + facet_wrap(~restaurant, ncol = 2) + geom_text(aes(label = signif(n, digits = 3)), nudge_y = 8) + labs(title = "Positive and Negative Words for Buca and Olive Garden", subtitle = "Only words appearing at least 10 times are shown") ``` The above shows that Buca consistently had positive words mentioned in their reviews. Olive Garden had many negative words included in their reviews. Recommend was mentioned many times by both however, we cannot assume that the 16 reviews, for olive garden, that say recommend were using the word in a negative or positive way. The last way to get an overall feel of the two restaurants is to categorize the words used in the reviews into characteristics. ```{r} #| label: Sentiment Graph #| include: TRUE nrc <- get_sentiments("nrc") tidy_reviews %>% inner_join(nrc, by = "word", relationship = "many-to-many") %>% group_by(sentiment, restaurant) %>% summarize(n = n()) %>% ggplot(aes(x = sentiment, y = n, fill = restaurant)) + geom_bar(stat = "identity", position = "dodge") + scale_fill_manual(values = c("#CB2147","#9FA73A")) + labs(title = "Restaurant Sentiment Scores", subtitle = "Total number of emotive words scored ", y = "Total Number of Words", x = "Emotional Sentiment", fill = "Restaurant") ``` The above shows that Olive Garden could be reviewed with sentiments of trust and joy. However they also have more sentiments of anger, disgust, fear, and sadness. Buca has more sentiments of anticipation and surprise in their reviews. ## Concluding Thoughts Looking at the sentiment analysis of Olive Garden and Buca Di Beppo there is positive sentiment surrounding the reviews for Buca and a negative sentiment for Olive Garden. Olive Gardens' reviews had word pairs that included wait, and they have many negative words included in their reviews. Olive Garden also was considered horrible and had significantly more negative reviews than Buca.