2026-09-10 –, Auditorium
Some time ago, I worked with a dataset containing textual data. At that time, I performed entropy calculations using Microsoft Excel. In this study, however, I aim to explore the use of LibreOffice Calc for data processing. Using the same dataset, I conducted sentiment labeling, topic relevance analysis, and entropy calculations for each column. The entropy calculation was performed to determine the weight value of each word.
The first step was to identify the number of unique words appearing in the dataset. After collecting all unique words, the next step was to calculate the frequency of each word. Once the frequency values had been obtained, the subsequent step was to calculate the probability of occurrence for each word. An example of this process is presented below:
Unique Word Frequency Probability
Train 2 2/4 = 0.50
Fire 1 1/4 = 0.25
Bekasi 1 1/4 = 0.25
In LibreOffice Calc, the formula used to calculate the entropy contribution of a probability value is:
=-(0.5*LN(0.5)/LN(2))
The result of this calculation is 0.5. This serves as an example of entropy calculation based on a probability value. In information theory, entropy is used to measure the amount of uncertainty or information contained in a dataset. The same calculation can be applied to other probability values, and the resulting entropy contributions can then be summed to obtain the total entropy of the dataset.
In addition, I calculated the probabilities of sentiment categories that had previously been grouped into three classes: positive, negative, and neutral. The first step was to determine the number of instances belonging to each sentiment category. To perform this calculation in LibreOffice Calc, I used the following formulas:
=COUNTIF(A:A,"positive")
=COUNTIF(A:A,"negative")
=COUNTIF(A:A,"neutral")
These formulas count the occurrences of each sentiment label within the dataset. The resulting counts are then used to calculate the probability of each sentiment category by dividing the frequency of each category by the total number of records. The results are presented in the following table:
A B
Sentiment Frequency
Positive 140
Negative 35
Neutral 16
Total 191
After obtaining the counts for the positive, negative, and neutral sentiment categories, the next step was to calculate the probability of each sentiment. Before calculating these probabilities, the total number of sentiment instances was determined by summing all sentiment categories. Subsequently, the probability of each sentiment was calculated by dividing the frequency of a sentiment category by the total number of records.
For example, the formula used in LibreOffice Calc was:
=B2/$B$5
Using this formula, the probability of the positive sentiment category was calculated as 0.73298. The same procedure was applied to the negative and neutral sentiment categories. The complete results are presented in the following table:
A B C
Sentiment Category Frequency Probability
Positive 140 0.73298
Negative 35 0.18325
Neutral 16 0.08377
Total 191
LibreOffice Calc proved to be a valuable alternative for this study due to its ease of use and user-friendly interface. The software was straightforward to understand and provided a comfortable working environment for data processing and analysis. Based on this experience, I would recommend the use of LibreOffice Calc to other students and researchers who require an accessible and effective tool for data analysis.
My name is Nawindah, I am a Lecturer at Budi Luhur University. The courses I teach are Analysis and Design Algorithm, Software project management. The product of my community service is the "Smart Senyum" mental health early detection application. My research area is mental health and Fuzzy Logic.
Sibro Abdurrahman is an undergraduate Computer Science student with research interests in Data Mining, Data Science, and Artificial Intelligence. His work focuses on text mining, sentiment analysis, and machine learning applications. He is passionate about leveraging data-driven methods to solve real-world problems and continuously expands his knowledge through academic research and practical projects.