Seleksi Fitur Bobot Kata dengan Metode TFIDF untuk Ringkasan Bahasa Indonesia

Widyasanti, Ni Komang; Putra, I Ketut Gede Darma; Rusjayanthi, Ni Kadek Dwi

doi:10.24843/jim.2018.v06.i02.p06

Cited by 15 publications

(13 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Metode TF-IDF juga dapat meningkatkan akurasi dalam proses analisis. Proses penghitungan bobot kata pada metode TF-IDF merupakan hasil dari penggabungan dua konsep yaitu TermFrequence (TF) dan Inverse Document Frequency (IDF) [15]. Proses penghitungan pembobotan TF-IDF menggunakan modul scikit-learn, yaitu TfidfVectorizer().…”

Section: Klasifikasi Naïve Bayesunclassified

Analisis Sentimen Customer Feedback Tokopedia Menggunakan Algoritma Naïve Bayes

Ama¹,

Mulya²,

Astuti³

et al. 2022

json

View full text Add to dashboard Cite

Products and customers have a close relationship, therefore UMKM need to build good relationships with customers. The most common way that companies or UMKM do is to look at the reviews given, this is called customer feedback. The results of customer feedback to companies or UMKM can improve service and product quality. The problem that arises is how to process the many reviews given, especially reviews from marketplaces like Tokopedia. Therefore, a method is needed to see user reviews of the products being sold, whether positive or negative. The method that will be used is sentiment analysis. Sentiment analysis is the process of understanding and extracting and automatically processing text data and can produce sentiments that are displayed in a sentence. The steps taken were taking House of Smith customer review data at Tokopedia, manual labeling to get positive and negative data reviews, data preprocessing, TF-IDF weighting and classification using the Naïve Bayes algorithm. The results of sentiment testing using the Naïve Bayes algorithm with TF-IDF weighting quality accuracy of 83% with visualization of the distribution of words that appear the most are the words 'good', 'comfortable' and 'use' for positive reviews. The most frequent negative reviews were 'material' and 'thin' which indicated that some buyers felt that the product had a thin material.

show abstract

Section: Klasifikasi Naïve Bayesunclassified

Analisis Sentimen Customer Feedback Tokopedia Menggunakan Algoritma Naïve Bayes

Ama¹,

Mulya²,

Astuti³

et al. 2022

json

View full text Add to dashboard Cite

show abstract

“…Term Frequency-Inverse Document Frequency atau TF-IDF merupakan metode yang berguna untuk menghitung bobot setiap kata yang umum digunakan. Metode ini banyak digunakan karena efisien, mudah dan memiliki hasil yang akurat seperti pada [8]- [10]. Metode ini menghitung nilai Term Frequency (TF) dan Inverse Document Frequency (IDF) pada setiap token (kata) di setiap dokumen dalam korpus.…”

Section: Metode Penelitianunclassified

Analisis Sentimen Pada Media Sosial Twitter Terhadap Pelayanan Dinas Kependudukan Dan Pencatatan Sipil Menggunakan Algoritma Naïve Bayes

Salim

Solichin

2022

IDEALIS

View full text Add to dashboard Cite

Dinas Kependudukan dan Pencatatan Sipil (Disdukcapil) memiliki tugas utama dalam memberikan layanan kependudukan dan pencatatan sipil bagi masyarakat. Saat ini, di kalangan masyarakat masih sering memiliki stigma kurang baik terhadap proses pengurusan dokumen kependudukan yang terkesan lama, rumit dan sering terjadi pungutan liar oleh pihak-pihak tertentu. Pada penelitian ini dilakukan analisis sentimen masyarakat terhadap layanan Disdukcapil pada media sosial Twitter. Analisis sentimen diperoleh menggunakan pendekatan machine learning disertai dengan ekstraksi fitur menggunakan TF-IDF dan algoritma Naïve Bayes. Dataset yang digunakan pada penelitian ini diperoleh dari media sosial Twitter berupa kicauan (tweet) berbahasa Indonesia. Dataset diperoleh menggunakan kata kunci ‘dukcapil’, ‘kependudukan’, ‘disdukcapil’, ‘ngurus ktp’, ‘pembuatan ktp’, ‘pengalaman ktp’, ‘pelayanan disdukcapil’, ‘ngurus kk’, ‘pelayanan ktp’, ‘ktp rusak’, ‘dokumen kependudukan’, ‘ngurus disdukcapil’, ‘bikin kartu keluarga’, ‘pelayanan dukcapil’, ‘bikin akta lahir’, ‘buat akta lahir’, dan ‘disdukcapil ramah’. Hasil analisis berdasarkan 647 tweet menunjukan bahwa sentimen positif sebesar 70,20% dan sentimen negatif sebanyak 29,80%, dengan nilai akurasi 65,39% presisi 97,65%, recall 66,87%, dan f1-score sebesar 39,69%. Hasil penelitian ini bermanfaat bagi Disdukcapil untuk meningkatkan kualitas layanannya sehingga sentimen masyarakat semakin baik. ----------------------------------------- The Department of Population and Civil Registration (Disdukcapil) has the main task of providing population and civil registration services for the community. Currently, the community still often has a bad stigma about the process of managing population documents which seems long, complicated and often illegal levies occur by certain parties. In this study, an analysis of public sentiment towards the Disdukcapil service was carried out on Twitter social media. Sentiment analysis was obtained using a machine learning approach accompanied by feature extraction using TF-IDF and the Naïve Bayes algorithm. The dataset used in this study was obtained from Twitter social media in the form of Indonesian-language tweets. The dataset was obtained using the keywords 'dukcapil', 'population', 'disdukcapil', 'ngurus ktp', 'ktp making', 'ktp experience', 'disdukcapil service', 'take care of family card', 'ktp service', 'broken ID card', 'residency document', 'manage disdukcapil', 'make a family card', 'dukcapil service', 'make a birth certificate', 'make a certificate born', and 'friendly disdukcapil'. The results of the analysis based on 647 Tweets show that positive sentiment is 70.20% and negative sentiment is 29.80%, with an accuracy value of 65.39%, precision of 97.65%, recall of 66.87, and f1-score of 39.69%. The results of this study are useful for Disdukcapil to improve the quality of its services so that public sentiment is getting better.

show abstract

“…There are many methods for this, but the popular one is called TF-IDF ("Term Frequency -Inverse Document" Frequency) which assigns a score to each word. TF-IDF is used directly from the word cloud generator in Python where in general TF-IDF is to summarize how often a word appears in a document (TF) and provide scaling (in descending order) to words that appear in a document (IDF) [14] . So the TF-IDF formulation is not defined by the researcher but directly uses the word cloud generator in Python which contains the TF-IDF formulation.…”

Section: Word Vectorization With Tf-idfmentioning

confidence: 99%

Word Cloud of UKSW Lecturer Research Competence Based on Google Scholar Data

Trihandaru

Parhusip

Susanto

et al. 2021

khif

View full text Add to dashboard Cite

There is a need in the Universitas Kristen Satya Wacana (UKSW) to identify the research competence of their faculties at a study program and University level. To accomplish this requirement, we need to automate the analysis of research output and publications quickly. Research articles are scattered in many publisher systems and journals which may be reputable, unreputable, accredited, and unaccredited. We devise a computer code to quickly and efficiently retrieve publication titles recorded in Google Scholar using a machine learning algorithm. The result display is in the form of a word cloud so that dominant and frequent words will be prominent in the visualization. In determining scientific terms to display, we used a modified version of the word cloud Python module and unmodified Term Frequency -Inverse Document Frequency (TF-IDF) library. The algorithm was tested on publication titles of our study program in UKSW and confirmed directly. The system features the ability to produce a word cloud visualization for an individual faculty, for faculties in a study program, or in the University as a whole. We have not differentiated publication sources, whether they are reputable or unreputable, which might affect the accuracy of competence identification.

show abstract

Seleksi Fitur Bobot Kata dengan Metode TFIDF untuk Ringkasan Bahasa Indonesia

Cited by 15 publications

References 5 publications

Analisis Sentimen Customer Feedback Tokopedia Menggunakan Algoritma Naïve Bayes

Analisis Sentimen Customer Feedback Tokopedia Menggunakan Algoritma Naïve Bayes

Analisis Sentimen Pada Media Sosial Twitter Terhadap Pelayanan Dinas Kependudukan Dan Pencatatan Sipil Menggunakan Algoritma Naïve Bayes

Word Cloud of UKSW Lecturer Research Competence Based on Google Scholar Data

Contact Info

Product

Resources

About