Customer Opinion Analysis

1. Load and Clean Data

# قراءة الملف وتنظيف أسماء الأعمدة
data <- read_excel("D:/R analysis/customer_opinions (1).xlsx") %>%
  clean_names()

# عرض أول الصفوف
head(data)

## # A tibble: 6 × 9
##      id company review_source username           review_date         review_text
##   <dbl> <chr>   <chr>         <chr>              <dttm>              <chr>      
## 1     1 Amazon  trustpilot    Deborah Stevenson… 2025-07-02 00:00:00 "Absolutel…
## 2     2 Amazon  trustpilot    Rui Oliveira       2025-07-02 00:00:00 "Don’t w…
## 3     3 Amazon  trustpilot    chris schoepf      2025-06-27 00:00:00 "Amazon co…
## 4     4 Amazon  trustpilot    Mokshya Priyadars… 2025-07-04 00:00:00 "Amazon Pr…
## 5     5 Amazon  trustpilot    SG                 2025-07-04 00:00:00 "Customer …
## 6     6 Amazon  trustpilot    customer           2025-07-03 00:00:00 "Don’t r…
## # ℹ 3 more variables: star_rating <dbl>, sentiment <chr>, category <chr>

2. Data Overview

# ملخص إحصائي شامل
skim(data)

Data summary

Name	data
Number of rows	22
Number of columns	9
_______________________
Column type frequency:
character	6
numeric	2
POSIXct	1
________________________
Group variables	None

Variable type: character

skim_variable	n_missing	complete_rate	min	max	empty	n_unique	whitespace
company	0	1.00	6	6	0	1	0
review_source	0	1.00	10	10	0	1	0
username	0	1.00	2	27	0	22	0
review_text	0	1.00	54	1339	0	22	0
sentiment	0	1.00	8	13	0	3	0
category	1	0.95	13	28	0	12	0

Variable type: numeric

skim_variable	n_missing	complete_rate	mean	sd	p0	p25	p50	p75	p100	hist
id	0	1	11.50	6.49	1	6.25	11.5	16.75	22	▇▆▆▆▇
star_rating	0	1	1.36	0.90	1	1.00	1.0	1.00	5	▇▂▁▁▁

Variable type: POSIXct

skim_variable	n_missing	complete_rate	min	max	median	n_unique
review_date	0	1	2025-06-27	2025-07-06	2025-07-03 12:00:00	6

3. Star Rating Analysis

# توزيع التقييمات
ggplot(data, aes(x = star_rating)) +
  geom_bar(fill = "steelblue") +
  labs(title = "Star Rating Distribution", x = "Stars", y = "Number of Reviews")

4. Sentiment Analysis

# عدد المراجعات حسب المشاعر
data %>% count(sentiment, sort = TRUE)

## # A tibble: 3 × 2
##   sentiment         n
##   <chr>         <int>
## 1 Very Negative    17
## 2 Negative          4
## 3 Very Positive     1

# رسم بياني للمشاعر
ggplot(data, aes(x = sentiment)) +
  geom_bar(fill = "darkgreen") +
  labs(title = "Sentiment Distribution", x = "Sentiment", y = "Number of Reviews")

5. Category Analysis

# توزيع الفئات
data %>%
  count(category, sort = TRUE) %>%
  ggplot(aes(x = reorder(category, n), y = n)) +
  geom_bar(stat = "identity", fill = "purple") +
  coord_flip() +
  labs(title = "Review Count by Category", x = "Category", y = "Number of Reviews")

6. Text Analysis

Most Frequent Words

# استخراج الكلمات الشائعة من التعليقات
words <- data %>%
  select(review_text) %>%
  unnest_tokens(word, review_text) %>%
  anti_join(stop_words) %>%
  count(word, sort = TRUE)

# عرض أول 10 كلمات
head(words, 10)

## # A tibble: 10 × 2
##    word           n
##    <chr>      <int>
##  1 amazon        24
##  2 customer      16
##  3 â             16
##  4 service       14
##  5 delivery      11
##  6 prime          9
##  7 time           9
##  8 donâ           7
##  9 experience     7
## 10 delivered      6

Word Cloud

# رسم سحابة الكلمات
wordcloud(words = words$word, freq = words$n, min.freq = 5,
          max.words = 100, random.order = FALSE, 
          colors = brewer.pal(8, "Dark2"))

7. Conclusions & Recommendations

• The average rating reflects overall satisfaction.
• Negative sentiment is concentrated in some categories.
• Frequent keywords show recurring issues (e.g., delay, quality, support).
• Recommend deeper analysis by company and time trends.