Transparent Aspect-Level Sentiment Analysis Based on Dependency Syntax Analysis and Its Application on COVID-19

2022 ◽  
Vol 14 (2) ◽  
pp. 1-24
Bin Wang ◽  
Pengfei Guo ◽  
Xing Wang ◽  
Yongzhong He ◽  
Wei Wang

Aspect-level sentiment analysis identifies fine-grained emotion for target words. There are three major issues in current models of aspect-level sentiment analysis. First, few models consider the natural language semantic characteristics of the texts. Second, many models consider the location characteristics of the target words, but ignore the relationships among the target words and among the overall sentences. Third, many models lack transparency in data collection, data processing, and results generating in sentiment analysis. In order to resolve these issues, we propose an aspect-level sentiment analysis model that combines a bidirectional Long Short-Term Memory (LSTM) network and a Graph Convolutional Network (GCN) based on Dependency syntax analysis (Bi-LSTM-DGCN). Our model integrates the dependency syntax analysis of the texts, and explicitly considers the natural language semantic characteristics of the texts. It further fuses the target words and overall sentences. Extensive experiments are conducted on four benchmark datasets, i.e., Restaurant14, Laptop, Restaurant16, and Twitter. The experimental results demonstrate that our model outperforms other models like Target-Dependent LSTM (TD-LSTM), Attention-based LSTM with Aspect Embedding (ATAE-LSTM), LSTM+SynATT+TarRep and Convolution over a Dependency Tree (CDT). Our model is further applied to aspect-level sentiment analysis on “government” and “lockdown” of 1,658,250 tweets about “#COVID-19” that we collected from March 1, 2020 to July 1, 2020. The experimental results show that Twitter users’ positive and negative sentiments fluctuated over time. Through the transparency analysis in data collection, data processing, and results generating, we discuss the reasons for the evolution of users’ emotions over time based on the tweets and on our models.

2021 ◽  
pp. 1-32
Magy Seif El-Nasr ◽  
Truong Huy Nguyen Dinh ◽  
Alessandro Canossa ◽  
Anders Drachen

This chapter introduces the topic of this book: Game Data Science. Game data science is the process of developing data-driven techniques and evidence to support decision-making across operational, tactical, and strategic levels of game development, and this is why it is so valuable. This chapter introduces this topic as well as outlines the process of game data science from instrumentation, data collection, data processing, data analysis, to reporting. Further, the chapter also discusses the application of game data science, as well as its utility and value, to the different stakeholders. The chapter also includes a section discussing the evolution of this process over time, which is important to situate the field and the techniques discussed in the book. The chapter also outlines established industry terminologies and defines their use in the industry and academia.

2020 ◽  
Vol 27 (1) ◽  
pp. 94
Ahmad Hasan Saifurrisal ◽  
Suyoto Suyoto ◽  
Sri Uchtiawati ◽  
Nur Fauziyah

Hasil pengamatan yang dilakukan di kelas XI-B SMA Plus Ar-Rahmat Bojonegoro menunjukkan bahwa sebagian besar peserta didik mengalami kesulitan pada materi polinomial. Kesulitan yang dialami peserta didik diduga karena guru lebih sering mengajar dengan model pembelajaran ekspositori sehingga peserta didik kurang aktif dalam proses belajar mengajar dan menyebabkan hasil belajar peserta didik kurang maksimal. Maka, perlu diterapkan metode pembelajaran yang dapat meningkatkan aktivitas dan hasil belajar peserta didik, yaitu discovery learning dengan strategi REACT. Penelitian ini dilaksanakan dengan tujuan untuk mendeskripsikan langkah-langkah pembelajaran discovery learning dengan strategi REACT yang dapat meningkatkan aktivitas dan hasil belajar peserta didik kelas XI-B SMA Plus ArRahmat Bojonegoro materi polinomial.  Penelitian ini merupakan Penelitian Tindakan Kelas (PTK). Data penelitian berupa daftar nilai peserta didik pada tes di tiap akhir siklus dan data hasil observasi yang didapatkan dari observer. Pengumpulan data dilakukan dengan teknik observasi, angket, lembar penilaian aktivitas peserta didik, tes, catatan lapangan, dan dokumentasi. Data yang terkumpul dianalisis secara kualitatif dan kuantitatif.  Berdasarkan hasil analisis tersebut, diperoleh dua kesimpulan sebagai berikut. Pertama, bahwa penerapan discovery learning dengan strategi REACT pada materi polinomial yang dapat meningkatkan aktivitas dan hasil belajar peserta didik kelas XI-B SMA Plus Ar-Rahmat Bojonegoro memiliki langkah-langkah: 1) guru memberikan apersepsi dan motivasi; 2) menerapkan discovery learning dengan strategi REACT dengan tahapan: stimulation, problem statement, data collection, data processing, dan verification; 3) peserta didikmengomunikasikan hasil diskusi kelompok kemudian bersama-sama guru membuat kesimpulan. Kedua, bahwa pembelajaran dengan discovery learning dengan strategi REACT meningkatkan aktivitas dan hasil belajar peserta didikkelas XI-B SMA Plus Ar-Rahmat Bojonegoro. Skor rata-rata aktivitas peserta didik pada siklus I sebesar 27,5 dengan kategori baik, tetapi masih ada poin pengamatan yang bernilai 2. Skor rata-rata aktivitas peserta didik pada siklus II sebesar 35,5 dengan kategori sangat baik, dan setiap poin pengamatan mencapai minimal nilai 3. Hasil tes peserta didik pada siklus I menunjukkan bahwa 60% peserta didik yang memperoleh nilai mencapai KKM dengan rata-rata hasil tes yang diperoleh adalah sebesar 78,25. Hasil tes peserta didik pada siklus II menunjukkan bahwa 90% peserta didik yang memperoleh nilai mencapai KKM dengan rata-rata hasil tes yang diperoleh adalah sebesar 85,45.   

2019 ◽  
Vol 3 (1) ◽  
pp. 128
Syamsul Huda ◽  
Heikal Muhammad Zakaria Hakim

The number of tobacco manufacturing companies is decreasing significantly, above 60 (sixty) companies on average every year. However, the cigarette companies contribute a significant amount to the state revenues, with an increase of more than 5 (five) trillion rupiah every year. Based on the data, we want to know the feasibility of investing in a cigarette company in the future. The method used includes four stages starting from data collection, data analysis, preliminary data processing, and concluding. This research use three sample companies: PT. Bentoel Internasional Investama Tbk, PT. Gudang Garam Tbk, and PT. HM Sampoerna Tbk. The results showed that based on the ROI, NPV, IRR, and BEP, PT. Gudang Garam Tbk. is the most feasible, followed by PT. HM Sampoerna Tbk. Meanwhile, PT. Bentoel Internasional Investama Tbk not feasible for investment.

2021 ◽  
Vol 11 (2) ◽  
pp. 125
Chacuk Tri Sasongko ◽  
Nini Susanti Susanti

<p>Javanese literary works, especially the Panji tales, often feature human characters with animal names, such as Kuda Narawangsa, Kebo Kanigara, and Kidangwalangka. This naming phenomenon can also be found in old Javanese inscriptions. Recent studies generally concluded that such naming tradition occurred during the Kadiri-Majapahit era, and this was closely related to the banner of the army and the identity of <em>makasirkasir</em>. This study aims to reveal the motivation behind the naming tradition and the relationship between personal name, social status, and occupation of the person so named throughout the ancient Javanese era. This study uses Nyström’s onomastical approach, especially the concept of anthroponomics, namely the presuppositional meanings of proper names consisting of categorial, associative, and emotive meanings. This research utilized archaeological methods which involved data collection, data processing, and interpretation. Results show that this naming phenomenon was generally motivated by people’s appreciation of certain animals that had a special place and played an important role in the ancient Javanese society and culture. The correlation between the names and the characters’ social status and occupation has been found to be influenced by the sociocultural development during the Ancient Mataram and Kadiri-Majapahit periods.</p>

Handrie Noprisson ◽  
Marissa Utami

The development of the online travel booking business must be followed by increased revenue to maintain the trust of business partners such as investors and merchants. Increased purchases by consumers can be supported by several factors, one of which is purchase intention from a consumer perspective. This study looks at how the influence of a consumer perspective based on trust, perceived value, brand image, product diagnosicity on increasing purchase intention for online travel booking applications. Stages of research include data collection, data processing using SMARTPLS, interpretation of data processing results and drawing research conclusions. As a result of the study, the trust factor has the influence on purchase intentions with a t-value of 8.280. The perceived value influences purchase intentions in the online travel booking application with a t-value of 7.091. Brand image also has an influence on purchase intention with a score of 5.253.

EduFisika ◽  
2020 ◽  
Vol 5 (02) ◽  
pp. 131-139
Rosinta Dehong ◽  
Melkyanus Bili Umbu Kaleka ◽  
Ana Silfiani Rahmawati

This study aims to analyze the steps of applying the discovery learning model in physics learning. This type of research is qualitative with the type of case study. The subjects in this study were physics teachers and students of class X SMKN 2 Ende, totaling 33 students. The object of this research is the application of the discovery learning model. Data collection instruments using questionnaires and documentation. Data were analyzed descriptively. The results showed that the application of the discovery learning model in physics learning for class X students of SMK Negeri 2 Ende was in accordance with the syntax. The syntax of the discovery learning model consists of stimulation, problems, data collection, data processing, verification, generalization.

2021 ◽  
Vol 9 (1) ◽  
pp. 312
Alef Experancio Ximenes Dasilelo ◽  
Emilya Kalsum ◽  
Bontor Jumaylinda Br. Gultom

Landak Regency is one of the areas in West Kalimantan that is rich in cultural heritage. Various forms of cultural heritage come from 2 large ethntic, it is Dayak and Malay ethnic. Nevertheless, all the wealth and cultural heritage has not been integrated and has a special container. From the potential and problems, it is necessary to hold a container that can shelter and accommodate all components of cultural activities in the landak regency. This refers to the Cultural Center of Landak Regency. The design method used in the design of the Cultural Center is a seven-step design method, which starts from the search for ideas, data collection, data processing, data analysis, determining concepts and deciding the design results. This method refers to an area-scale Cultural Center design whose all variants of activities will be sheltered by the composition of several building times implemented based on their functions and activities, this Cultural Center will be supported by the application of a sheath system that will be implemented through the typical patterns of culture of Landak Regency with the emphasis of the concept more to modern minimalist buildings but rich in cultural value.

Jayashree K. ◽  
Abirami R. ◽  
Rajeswari P.

The successful development of big data and the internet of things (IoT) is increasing and influencing all areas of technologies and businesses. The rapid increase of more devices that are connected to IoT from which enormous amount of data are consumed indicates the way how big data is related with IoT. Since huge amount of data are obtained from different sources, analysis of these data involves much of processing at each and every level to extract knowledge for decision making process. To manage big data in a continuous network that keeps expanding leads to few issues related to data collection, data processing, analytics, and security. To address these issues, certain solution using bigdata approach in IoT are examined. Combining these two areas provides several opportunities developing new systems and identify advanced techniques to solve challenges on big data and IoT.

2020 ◽  
Vol 12 (3) ◽  
pp. 43-56 ◽  
Sukhnandan Kaur Johal ◽  
Rajni Mohana

Various natural language processing tasks are carried out to feed into computerized decision support systems. Among these, sentiment analysis is gaining more attention. The majority of sentiment analysis relies on the social media content. This web content is highly un-normalized in nature. This hinders the performance of decision support system. To enhance the performance, it is required to process data efficiently. This article proposes a novel method of normalization of web data during the pre-processing phase. It is aimed to get better results for different natural language processing tasks. This research applies this technique on data for sentiment analysis. Performance of different learning models is analysed using precision, recall, f-measure, fallout for normalize and un-normalize sentiment analysis. Results shows after normalization, some documents shift their polarity i.e. negative to positive. Experimental results show normalized data processing outperforms un-normalized data processing with better accuracy.

Multilingual ◽  
2019 ◽  
Vol 18 (1) ◽  
pp. 22-32
Rissari Yayuk

The research that will be raised is the metonymy of the "speaking" in the Banjar language in a traditional store (warung). Issues discussed included how the metonym structure of the "speaking" in Banjar was in store? The purpose of this study is to describe the metonym of the Banjar language "conversation" in the warung. The method used is descriptive method with a semantic approach. Data collection is done through reference and recording techniques. The author takes three steps of work, namely the stage of data collection, data processing, and the stage of presenting the results of data analysis. The data analysis technique is the distribution method. The presentation of data analysis describes conversations containing metonymy in Banjar language. Data presentations are written in ordinary words. The population of this study is the Banjar community located in the Gambah neighborhood, South Hulu Sungai Regency, South Kalimantan Province. The time of data collection is from January 2017 to March 2017. The results of this study include metonymy based on part elements with the whole, metonymy based on place attributes, metonymy based on objects for content or function and metonymy based on time attributes.Penelitian yang akan diangkat adalah metonimi panderan “pembicaraan” di warung bahasa Banjar. Masalah yang dibahas meliputi bagaimana struktur  metonimi panderan “pembicaraan” berbahasa Banjar di warung? Tujuan penelitian ini adalah mendeskripsikan struktur metonimi panderan “pembicaraan” di warung bahasa Banjar. Metode yang digunakan adalah metode deskriptif dengan pendekatan semantik. Pengumpulan data dilakukan melalui teknik simak dan catat. Penulis menempuh tiga langkah kerja, yaitu tahap pengumpulan data, pengolahan data, dan tahap penyajian  hasil analisis data. Teknik analisis data adalah metode agih. Sajian analisis data mendeskripsikan ujaran yang mengandung makna metonimi dalam bahasa Banjar. Penyajian data ditulis dengan kata-kata biasa. Populasi penelitian ini adalah masyarakat Banjar yang berlokasi  di lingkungan Gambah, Kabupaten Hulu Sungai Selatan, Provinsi Kalimantan Selatan.Waktu pengambilan data dari bulan Januari 2017 sampai dengan bulan Maret 2017. Hasil penelitian meliputi metonimi berdasarkan unsur bagian dengan keseluruhan, metonimi berdasarkan atribut tempat, metonimi berdasarkan objek untuk isi/fungsi dan metonimi berdasarkan atribut waktu.

Sign in / Sign up

Export Citation Format

Share Document