Using Rstudio write data sets and codes for this text mining project using these procedures;

1. write codes to install tm packages and set up directory.

2. write codes to read the data and check

3. write codes to convert charactor factor and check the project

4. write codes to creat corpus and clean it for the project.

5. write codes to clean corpus for the project.

6. write codes to compare data on the project for the processing.

NOTE; Below is the project to use these procedure to work on-

The purpose of this project is to explore the use of big data techniques, specifically text mining, to understand customer preferences and enhance business performance. With the increasing availability of data, businesses are now able to gather large amounts of information about their customers, including their preferences, opinions, and feedback. However, the challenge lies in making sense of this data and using it to make informed business decisions. Big data are now being collected by information technology (IT) devices, computer software, and search engine websites. Big data analytics refers to the process of analyzing big data to gain more insights for decision making (APICS, 2012). The project discusses how emerging information technologies (IT) such as big data analytics is being used to gather, process, and analyze data on customers for better understanding and insight into customer preference and feedback. By gaining a good insight on customer preference and feedback; companies and businesses are able to develop strategies to enhance customer service, boost customer satisfaction and reduce customer churn. The project aims to address this challenge by using text mining to extract insights from customer data and provide actionable recommendations to retain customers and reduce customer churn. The discussion is structured into four sections and sub-sections. The first section discusses the problem at hand that is faced by the business. The second section reviews up-to-date literature on big data analytics and their techniques. The third section discusses the application of analytical techniques to gain understanding of data on customer feedback, provides actionable insights, and discusses scenarios for implementation of solutions. The fourth and last section discusses the summary of the findings by comparing and contrasting with literature, the recommendations, and conclusion.

PROBLEM SCENARIO

In this section, it is discussed the problem at hand that is facing the business in question. Businesses face many challenges as they strive to grow, survive, and make profit. The challenges may be associated with customer dissatisfaction, customer churn, fall in market share, stiff competition, declining sale revenue, high employee turnover, etc.

The specific problem that is to be addressed is that of customer churn in the retail business. Customer churn refers to the loss of customers over time, which can have a significant impact on a business's revenue and profitability. In particular, the objective is to understand the reasons behind customer churn and identify ways to retain customers and improve their experience.

The problem scenario that is being attempted to address through this project is to understand customer preferences and enhance business performance for a local restaurant. The restaurant had been experiencing a decline in business, and the owner believed that understanding customer feedback could help improve the business. Big data technique, specifically text mining, would be used to analyze customer feedback and gain insights into customer preferences.

LITERATURE REVIEW

In this section, it is conducted a critique of up-to-date literature to understand the concept of big data and its techniques.

A comprehensive literature review was conducted on the use of text mining in customer analytics. It was found that text mining techniques can be used to extract insights from various sources of customer data, including customer feedback, social media posts, and online reviews. These insights can be used to improve customer satisfaction, increase customer retention, and ultimately drive business growth.

Text mining

Text mining is among various techniques used in data mining. Text mining refers to the process that requires extracting vital and useful information from unstructured text data (Berry, 2004). This data mining technique draws on methods and techniques used for information retrieval, statistics, machine learning, and computational linguistics (Berry, 2004). By this It been argued that text data constitute over 80% of all stored online data (Gupta and Lehal, 2009). This suggests therefore that text analysis and mining is of great importance to business analyst, data analyst, and data scientist. The problem with text data is that it is largely unstructured and therefore need to be analyzed in such a manner so as to make meaning out of them.

Text mining differs from data mining in multiple ways. Text mining focuses on making sense of unstructured or semi-structured data while data mining deals more with structured data. While text mining is appropriate for analyzing textual non-quantitative data, data mining is more appropriate for analyzing numeric data (Gupta and Lehal, 2009).

Text mining techniques are used in multiple disciplines such as information technology, telecommunications, finance (banking, insurance, investment), business, health, psychology, biology, etc. Texting mining is particularly useful for business analysis and can be applied to the area of marketing and customer relationship management, human resource management, and financial management, risk management,

Several analytical techniques that are commonly used in text mining including but not limited to information extraction, information retrieval, categorization, clustering, visualization, summarization, sentiment analysis, topic modeling, and entity extraction was discussed. Sentiment analysis is a technique used to determine the sentiment or opinion expressed in a piece of text, while topic modeling is used to identify the main themes or topics discussed in a large collection of text. Entity extraction, on the other hand, is used to identify specific entities mentioned in a piece of text, such as people, places, or products (Gupta and Lehal, 2009).

SECTION THREE: ANALYTICS, ACTIONABLE INSIGHTS, AND SCENARIOS

In this section, the discussion is about the application of analytical techniques to gain understanding of data on customer feedback, provides actionable insights, and discusses scenarios for implementation of solutions.

Data collection and analysis

Customer data with respect to preferences, tastes, satisfaction, and feedback were collected using company or product websites and social media platforms such as facebook, twitter, tiktok, etc. By conducting customer and market survey; the restaurant gathered data on customers' preferences, customers' tastes, customers' satisfaction, and customer feedback. Collected data were analyzed statistically and textually using text mining techniques such as sentiment analysis, topic modelling and entity extraction. Using the statistical software, RStudio, the sentiment analysis, topic modelling, and entity extraction were performed to make meaning of the gathered data.

Analytical Techniques

Using these techniques, it was conducted an analysis of customer feedback data collected from a retail business. Sentiment analysis technique was used to determine the overall sentiment of customer feedback, as well as the sentiment associated with specific products or services. Topic modeling technique was used to identify the main themes discussed in customer feedback, such as product quality, customer service, and pricing. Entity extraction technique was used to******

Using topic modelling technique and with the assistance of RStudio, frequency of occurrence was generated for each term as representing key attributes for customer service based on the customer feedback data that was uploaded into the software for statistical analysis. Table 1 presents the number of times key words relating to customer service and customer satisfaction are raised by customers of a local restaurant.

Table 1: Top 10 Keywords in Customer Feedback Data

Keyword	Frequency
Quality	1,236
Service	985
Price	875
Selection	654
Cleanliness	532
Atmosphere	453
Convenience	378
Location	276
Staff	187
Wait time	143

Source: Customer Survey Analysis (2023)

The list of key words as presented in Table 1 follows in order of number of times repeated by customers. From Table 1, the statistical results revealed that the term "quality" appears 1 236 times in customer feedback survey regarding key aspects which customers consider important to improving customer service and therefore ensuring customer satisfaction. The term "service" comes second as it appears 985 times in customer feedback survey on key attributes that customers perceive as important in the restaurant business. Price is ranked third as it appears 875 times in customer feedback survey. Selection ranked fourth (4th) as it appears 654 times in the customer feedback survey, Cleanliness ranked fifth (5th) as it appears 532 times in the customer feedback survey. Atmosphere ranked sixth (6th) as it appears 453 times in the customer feedback survey. Convenience ranked seventh (7th) as it appears 378 times in the customer feedback survey. Location ranked eighth (8th) as it appears 276 times in the customer feedback. Staff ranked ninth (9th) as it appears 187 times in the customer feedback survey. Wait time ranked tenth (10th) as it appears 143 times in the customer feedback survey. From these statistical results, it is observed that key attributes stressed by customers during the feedback survey were quality, service, and price as they rank in the first three attributes.

Using sentiment analysis and with the assistance of RStudio, the sentiment score was computed for sentimental statement expressed by customers. Table 2 presents sentiment score for customer feedback survey based on the customer feedback data. Sentiment score was high (0.85) for service and price as customers are of the opinion that service is excellent and prices are reasonable while customers' opinion differs when quality of products.

Table 2: Customer Feedback Data

Feedback	Sentiment Score
The service was excellent and the prices were reasonable.	0.85
The selection was good, but the restaurant was not very clean.	0.65
The quality of the products was poor and the staff was unhelpful.	-0.75

Source: Customer Survey Analysis (2023)

Table 3 presents the correlation results for key customer service attributes.

Table 3: Correlation Matrix of Customer Feedback Keywords

Quality

Service

Price

Selection

Cleanliness

Atmosphere

Convenience

Location

Staff

Wait Time

Quality	1.00	0.45	-0.23	0.57	0.38	0.32	0.24	0.15	0.18	-0.05
Service	0.45	1.00	0.12	0.65	0.51	0.46	0.39	0.28	0.62	0.21
Price	-0.23	0.12	1.00	-0.35	-0.26	-0.15	-0.20	-0.31	-0.08	0.13
Selection	0.57	0.65	-0.35	1.00	0.43	0.38	0.31	0.12	0.23	-0.10
Cleanliness	0.38	0.51	-0.26	0.43	1.00	0.49	0.36	0.21	0.29	-0.05
Atmosphere	0.32	0.46	-0.15	0.38	0.49	1.00	0.27	0.18	0.25	-0.08
Convenience	0.24	0.39	-0.20	0.31	0.36	0.27	1.00	0.34	0.21	-0.15
Location	0.15	0.28	-0.09	0.22	0.15	0.18	0.30	1.00	0.18	-0.13
Staff	0.18	0.62	-0.08	0.23	0.29	0.25	0.21	0.18	1.00	0.12
Wait Time	-0.05	0.21	0.13	-0.10	-0.05	-0.08	-0.15	-0.13	0.12	1.00

Source: Customer Survey Analysis (2023)

Note: The values in the table represent the Pearson correlation coefficient between the corresponding pairs of feedback keywords.

From Table 3, the correlation coefficients weak, moderate, and strong negative and positive relationship among key attributes. "Quality" has moderate positive association with Service (0.45) and Selection (0.57) suggesting Quality moves moderately in the same direction as Service and Selection; weak negative association with Price (0.23), Cleanliness (0.38), Atmosphere (0.32), Convenience (0.24), Location (0.15), and Staff (0.18) suggesting Quality moves slowly in the same direction as Price, Cleanliness, Atmosphere, Convenience, Location, and Staff; and weak negative association (-0.05) with Wait time suggesting Quality moves slowly in opposite direction as Wait time.

"Service" has moderate positive association with Selection (0.65), Cleanliness (0.51), Atmosphere (0.46), and Staff (0.62) suggesting Service moves moderately in the same direction as Selection, Cleanliness, Atmosphere, and Staff; and weak positive association with Price (0.12), Convenience (0.39), Location (0.28) and Wait time (0.21) suggesting Service moves slowly in the same direction as Price, Convenience, Location, and Wait time.

"Price" has weak positive association with Wait time (0.13) suggesting Price moves slowly with Wait time; and weak negative association with Selection (-0.35), Cleanliness (-0.26), Convenience (-0.15), Atmosphere (-0.20), Location (0.09), and Staff (-0.08) suggesting Service moves slowly in opposite direction as Selection, Cleanliness, Convenience, Atmosphere, Location, and Staff.

"Selection" has moderate positive association with Cleanliness (0.43) suggesting Selection moves moderately in the same direction as Cleanliness; weak negative association with Atmosphere (0.38), Convenience (0.31), Location (0.22), and Staff (0.23) suggesting Selection moves slowly in the same direction as Atmosphere, Convenience, Location, and Staff; and weak negative association with Wait time (-0.10) suggesting Service moves slowly in opposite direction as Wait time.

"Cleanliness" has moderate positive association with Atmosphere (0.49) suggesting Cleanliness moves moderately in the same direction with Atmosphere; weak positive association with Convenience (0.36), Location (0.15), and Staff (0.29) suggestion Cleanliness moves slowly in the same direction as Convenience, Location, and Staff; and weak negative association with Wait time (-0.05) suggesting Cleanliness moves slowly in opposite direction as Wait time.

"Atmosphere" has weak positive association with Convenience (0.27), Location (0.18), and Staff (0.25) suggestion Atmosphere moves slowly in the same direction as Convenience, Location, and Staff; and weak negative association with Wait time (-0.08) suggesting Atmosphere moves slowly in opposite direction as Wait time.

"Convenience" has weak positive association with Location (0.30), and Staff (0.21) suggestion Atmosphere moves slowly in the same direction as Location, and Staff; and weak negative association with Wait time (-0.15) suggesting Convenience moves slowly in opposite direction as Wait time.

"Location" has weak positive association with Staff (0.18) suggestion Location moves slowly in the same direction as Staff; and weak negative association with Wait time (-0.13) suggesting Cleanliness moves slowly in opposite direction as Wait time.

"Staff" has weak positive association with Staff (0.12) suggestion Wait time moves slowly in the same direction as Wait time.

Actionable Insights

Our analysis revealed several actionable insights for the retail business. For example, we found that customers were generally satisfied with the quality of the products but had concerns about the pricing and the level of customer service. We also identified specific products and services that were associated with negative sentiments, such as a particular clothing line and the returns process. Based on these insights, we recommended that the business focus on improving customer service and addressing the concerns raised by customers in relation to specific products and services.

Scenario after Implementing Solutions

After implementing the recommendations, the retail business saw an increase in customer satisfaction and a reduction in customer churn. Specifically, the business was able to address the concerns raised by customers and improve the overall customer experience. As a result, customers were more likely to return to the restaurant and make repeat purchases, leading to an increase in revenue and profitability.

SECTION FOUR: DISCUSSION AND CONCLUSIONS

In this section, the discussion is about the summary of the findings by comparing and contrasting with literature, the recommendations, and conclusion.

Summary discussion

In conducting the project, it was first conducted a literature review to identify relevant analytical techniques. Text mining was chosen as it is a powerful technique for extracting insights from large amounts of unstructured data such as customer reviews. It was then collected customer feedback from various online platforms such as Yelp, Google Reviews, and TripAdvisor. It was cleaned the data by removing irrelevant information, such as website links, and converted it into a structured format.

Next, it was used text mining techniques such as sentiment analysis and topic modeling to extract insights from the data. Sentiment analysis helped us understand the overall sentiment of the reviews, while topic modeling helped us identify the most frequently discussed topics in the reviews. It was also created a correlation matrix to identify any relationships between customer feedback keywords.

Based on the analysis, it was identified several key insights. Customers placed a high value on quality and service, while price and wait time were less important. They also valued cleanliness, atmosphere, and convenience. It was provided actionable insights to the restaurant owner, such as the importance of maintaining high levels of cleanliness and improving customer service.

After implementing the insights, the restaurant owner saw a significant improvement in business performance. Customer satisfaction increased, and the restaurant received more positive reviews. It was also recommended that the owner continue to collect and analyze customer feedback to maintain a competitive edge.

Overall, the solution involved using big data techniques to analyze customer feedback and gain insights into customer preferences. It was provided actionable insights to the restaurant owner, which led to an improvement in business performance.

Conclusion

In conclusion, the project demonstrates the value of using big data techniques, specifically text mining, to understand customer preferences and enhance business performance. By analyzing customer feedback data actionable insights and recommendations were identified for the restaurant business to improve its customer experience and reduce customer churn. The findings are consistent with the literature on the use of text mining in customer analytics and highlight the importance of using data-driven techniques to make informed business decisions.

This project has been logically written and is tightly connected to the topic of using big data techniques, specifically text mining, to understand customer preferences and enhance business performance. The literature review is comprehensive and critically evaluates the relevant sources. The analytical techniques used are appropriate and the insights provided are actionable. The project is well-referenced, and tables and figures are used where appropriate to support arguments. Therefore, the overall quality of this project is high.

Recommendation for managers and future studies

Based on the findings, it is recommend that managers in restaurant businesses consider using text mining to analyze customer feedback data and identify areas for improvement. By doing so, they can gain valuable insights into customer preferences and opinions, which can be used to improve the customer experience and ultimately drive business growth.

One limitation of the project is that it was only analyzed customer feedback data from one restaurant business. Therefore, the generalizability of the findings may be limited to this particular context. Future research could explore the use of text mining in other industries and contexts to determine the effectiveness of these techniques in improving customer experience and reducing customer churn.

Another limitation is that the analysis was based solely on customer feedback data. Future research could explore the use of text mining in combination with other sources of data, such as sales data and customer demographics, to gain a more comprehensive understanding of customer preferences and behaviors.

Finally, there is also the need for further research into the ethical implications of using customer data in this way. As businesses collect more and more data about their customers, it is important to consider how this data is being used and whether it is being used in a way that is fair and transparent. Future research could explore the ethical implications of using text mining and other big data techniques in customer analytics and provide guidelines for businesses to ensure that their use of customer data is ethical and responsible.