Sentiment words analysis.Rmd

---
title: "Sentiment words analysis"
output: word_document
---

```{r}
if (!require(tidytext)) {install.packages("tidytext")}
if (!require(ggplot2)) {install.packages("ggplot2")}
if (!require(dplyr)) {install.packages("dplyr")}
```

read zomato and swiggy files

```{r read files}
zomato <- readLines('data/zomato.txt')
swiggy <- readLines("data/swiggy.txt")
```

function to return bing sentiment words 

```{r}
get_positive_negative_word_counts <- function(corpus) {
  df = data_frame(text = zomato)  #create dataframe from corpus
  bing_sentiments = get_sentiments("bing")
  
  word_counts <- df %>%
    unnest_tokens(word, text) %>%
    inner_join(bing_sentiments) %>%
    count(word, sentiment, sort = TRUE) %>%
    ungroup()
  
  return (word_counts)
}

```

bing sentiments add positive and negative scores
Based on those score contribution 

```{r}
swiggy_word_counts <- get_positive_negative_word_counts(swiggy)
swiggy_word_counts %>%
  filter(n > 20) %>%
  mutate(n = ifelse(sentiment == "negative", -n, n)) %>%
  mutate(word = reorder(word, n)) %>%
  ggplot(aes(word, n, fill = sentiment)) +
  geom_bar(stat = "identity") +
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
  xlab("Words") +
  ylab("Count of words") +
  ggtitle("Words contributing towards Swiggy sentiment")
```


```{r}
zomato_word_counts <- get_positive_negative_word_counts(zomato)
zomato_word_counts %>%
  filter(n > 20) %>%
  mutate(n = ifelse(sentiment == "negative", -n, n)) %>%
  mutate(word = reorder(word, n)) %>%
  ggplot(aes(word, n, fill = sentiment)) +
  geom_bar(stat = "identity") +
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
  xlab("Words") +
  ylab("Count of words") +
  ggtitle("Words contributing towards Zomato sentiment")
```