Laptop_EDA.Rmd

---
output:
  html_document:
    theme: flatly
    highlight: pygments
    toc: true
    toc_float:
      collapsed: true
      smooth_scroll: true
    toc_depth: 4
    df_print: paged
    code_folding: show
---

```{r, include=FALSE}
library(tidyverse)
library(tidytext)
library(scales)
library(AER)
library(lubridate)

options(scipen = 20)

knitr::opts_chunk$set(warning = FALSE,
                      message = FALSE,
                      fig.align = "center",
                      fig.width = 8)

```

```{r}
files <- dir()

files <- files[str_detect(dir(), "all_laptop_reviews.+")]
```



```{r}
laptop_data <- map_dfr(files, read_csv)
```

```{r}

# Filtering out missed duplicated products

laptop_data <- laptop_data %>% 
  filter(!Product_Name %in% c(
    'Samsung - Galaxy Book Pro 360 15.6" AMOLED Touch-Screen Laptop - Intel Evo Platform Core i7 - 16GB Memory - 1TB SSD - Mystic Navy',
    'ASUS - 14.0" Laptop - Intel Celeron N4020 - 4GB Memory - 64GB eMMC - Star Black - Star Black',
    'ASUS - 14" Chromebook - Intel Celeron N3350 - 4GB Memory - 32GB eMMC - Silver',
    'Samsung - Galaxy 13.3" 4K Ultra HD Touch-Screen Chromebook - Intel Core i5 - 8GB Memory - 256GB SSD - Mercury Gray'
  ))
```

```{r}
laptop_data_filtered <- laptop_data %>%
  mutate(Total_Vote = Helpful + Unhelpful) %>% 
  filter(Total_Vote > 4) %>% 
  mutate(Helpfulness = Helpful / Total_Vote)

laptop_data_filtered <- laptop_data_filtered %>% 
  mutate(Rev_dup = duplicated(Review)) %>% 
  filter(Rev_dup == 0)
```


```{r}
laptop_data_filtered %>% 
  ggplot(aes(Helpfulness))+
  geom_histogram(color = "white")
```


```{r}
laptop_data_filtered %>% 
  ggplot(aes(Rating))+
  geom_bar()
```



```{r}
laptop_data_filtered %>% 
  ggplot(aes(Rating, Helpfulness))+
  geom_point()+
  geom_smooth(method = "loess")
```



```{r}
laptop_data_filtered <- laptop_data_filtered %>% 
  mutate(Doc_ID = row_number())
```



```{r}
laptop_data_filtered <- laptop_data_filtered %>% 
  mutate(Review = str_to_lower(Review, locale = "en"))
```



```{r}
words_agg <- laptop_data_filtered %>% 
  unnest_tokens(output = word, input = Review, to_lower = FALSE) %>% 
  anti_join(stop_words) %>% 
  count(Doc_ID)
```


```{r}
laptop_data_filtered <- words_agg %>% 
  inner_join(laptop_data_filtered, by = "Doc_ID") 
```


```{r}
laptop_data_filtered <- laptop_data_filtered %>% 
  mutate(Date = as_date(mdy_hm(Date, locale = "english"))) 
```

```{r}
first_dates <- laptop_data_filtered %>% 
  group_by(Product_Name) %>% 
  slice_min(order_by = Date, n = 1, with_ties = F) %>% 
  select(Product_Name, Date) %>% 
  rename(First_Date = Date)
```

```{r}
laptop_data_filtered <- full_join(laptop_data_filtered, first_dates, by = "Product_Name") %>% 
  mutate(Timeline = as.integer(Date - First_Date))
```


```{r}
laptop_data_filtered %>% 
  ggplot(aes(Timeline))+
  geom_histogram()

laptop_data_filtered %>% 
  ggplot(aes(Timeline, Helpfulness))+
  geom_point()+
  geom_smooth(method = "lm")
```


```{r}
laptop_data_filtered %>% 
  ggplot(aes(n))+
  geom_histogram(color = "white")+
  scale_x_log10()
```



```{r}
laptop_data_filtered %>% 
  ggplot(aes(n, Helpfulness))+
  geom_point()+
  geom_smooth()+
  facet_wrap(~Product_Name)+
  scale_x_continuous(n.breaks = 10)+
  coord_cartesian(ylim = c(0,1))
```




```{r}
laptop_data_filtered %>% 
  ggplot(aes(n, Rating))+
  geom_point()+
  geom_smooth()
```


```{r}
sent_scores <- laptop_data_filtered %>% 
  unnest_tokens(output = word, input = Review, to_lower = FALSE) %>% 
  inner_join(get_sentiments("afinn")) %>% 
  group_by(Doc_ID) %>% 
  summarise(Total_Sent = sum(value))
```



```{r}
laptop_data_filtered %>% 
  inner_join(sent_scores, by = "Doc_ID") %>% 
  ggplot(aes(Total_Sent, Helpfulness))+
  geom_point()+
  geom_smooth()
```


```{r}
laptop_data_filtered %>% 
  inner_join(sent_scores, by = "Doc_ID") %>% 
  ggplot(aes(Total_Sent, Rating))+
  geom_point()+
  geom_smooth()
```

```{r}
laptop_sents <- laptop_data_filtered %>% 
  inner_join(sent_scores, by = "Doc_ID")
```


```{r}
laptop_sents %>% 
  ggplot(aes(Total_Sent))+
  geom_histogram()
```

```{r}
laptop_data_filtered %>% 
  ggplot(aes(Total_Vote))+
  geom_histogram(color = "white")+
  scale_x_log10(breaks = scales::log_breaks(12))
```



```{r}
summary(lm(Helpfulness ~ Rating + I(Rating^2) + log10(n) + log10(Total_Vote) + Timeline,
           data = laptop_data_filtered))
```

```{r}
lm(Helpfulness ~ Rating + I(Rating^2) + log10(n) + log10(Total_Vote) + Timeline, data = laptop_data_filtered) %>% 
  autoplot(which = 1:6)
```


```{r}
laptop_data_filtered
```



```{r}
summary(tobit(Helpfulness ~ Rating + I(Rating^2) + log10(n)  + log10(Total_Vote) + Timeline,
              data = laptop_data_filtered,
              left = 0, right = 1))
```



```{r}
laptop_data_filtered %>% 
  mutate(n = log10(n)) %>% 
  select(Helpfulness, Rating, Total_Vote, n,Timeline) %>% 
  GGally::ggpairs()
```








```{r}
model <- tobit(Helpfulness ~ Rating + I(Rating^2) + log10(n) + log10(Total_Vote), data = laptop_data_filtered, left = 0, right = 1)
model2 <- tobit(Helpfulness ~ Rating + I(Rating^2) + I(Rating^3) + log10(n) + log10(Total_Vote), data = laptop_data_filtered, left = 0, right = 1)

summary(model)
summary(model2)

anova(model2)
```


```{r}
summary(model)

preds <-  predict(model2, newdata = tibble(Rating = seq(0.5, 5, 0.5),
                                    n = rep(100,10),
                                    Total_Vote = rep(50,10))) 

qplot(x = seq(0.5, 5, 0.5), preds)
```