A Text Mining Based Study of Investor Sentiment and Its Influence on Stock Returns

Shi, Yong; Tang, Ye-ran; Cui, Ling-Xiao; Long, Wen

doi:10.24818/18423264/52.1.18.11

Cited by 19 publications

(11 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Note that dr-ml is the abbreviation of dimensionality reduction and machine learning Also, random forest and gradient boosting are abbreviated by rf and gb, respectively [20] to more extensive, for example deterministic, classes of frameworks. It is also desirable to apply the sparse PCA to other areas of text analysis and data mining [21][22][23][24] and particularly big data [25].…”

Section: Discussionmentioning

confidence: 99%

Sparse Principal Component Analysis for Natural Language Processing

Drikvandi

Lawal

2020

Ann. Data. Sci.

View full text Add to dashboard Cite

High dimensional data are rapidly growing in many different disciplines, particularly in natural language processing. The analysis of natural language processing requires working with high dimensional matrices of word embeddings obtained from text data. Those matrices are often sparse in the sense that they contain many zero elements. Sparse principal component analysis is an advanced mathematical tool for the analysis of high dimensional data. In this paper, we study and apply the sparse principal component analysis for natural language processing, which can effectively handle large sparse matrices. We study several formulations for sparse principal component analysis, together with algorithms for implementing those formulations. Our work is motivated and illustrated by a real text dataset. We find that the sparse principal component analysis performs as good as the ordinary principal component analysis in terms of accuracy and precision, while it shows two major advantages: faster calculations and easier interpretation of the principal components. These advantages are very helpful especially in big data situations.

show abstract

Section: Discussionmentioning

confidence: 99%

Sparse Principal Component Analysis for Natural Language Processing

Drikvandi

Lawal

2020

Ann. Data. Sci.

View full text Add to dashboard Cite

show abstract

“…Hasil kajian ini menunjukkan bahwa selain berpartisipasi pada komunitas virtual publik, mereka juga berpartisipasi pada jejaring sosial private yang memberi benefit berupa informasi dan pengetahuan baru. Kemudian text mining (ekstraksi dan klasifikasi) untuk menganalisis sentimen investor bursa efek China atas return saham dalam Index CSI 300 menunjukkan hasil bahwa terdapat faktor pengaruh akumulasi sentimen pada return harga saham (Shi, 2019). Sedangkan efek penularan rumor dan meningkatnya risiko investor dalam keterkaitannya dengan harga saham disimulasikan menggunakan SCIR model dalam jaringan bi-layer berpasangan sehingga diketahui dampaknya baik secara global maupun lokal (Dong, 2019 dasarnya mengandung risiko.…”

Section: Pendahuluanunclassified

Prediksi Sentimen Investor Pasar Modal Di Jejaring Sosial Menggunakan Text Mining

Mahani

Margono

2021

BLC

View full text Add to dashboard Cite

The decline in optimism for capital market investors is one of the financial impacts on the business world that arose from the SARS-COVID19 pandemic. This event was reflected in a decrease in trading volume followed by a sharp drop in the JCI on the Indonesia Stock Exchange starting March 2020. Thus, a slowdown in the economic recovery resulting from the pandemic is reflected in investor sentiment in the capital market. On the one hand, the rapid development of the internet in Indonesia has triggered the investor's activities in the information searching prior buy and sell securities, mostly use online platforms, which contribute to influencing investor preferences and sentiment. This study conducted a qualitative examination of the features/terms of stock investment in the capital market and collected them in a compact dictionary (lexicon). Therefore, lexicon-based investor opinion extraction was extracted from Twitter, followed by the text sentiment analysis, and forming a classification model based on Naive Bayes and Decision Tree. This research output shows that the polarity of capital market investor sentiment is optimistic with the sentiment features that often appear, namely "cuan", "bearish," "serok", "copet", "untung", "cut loss", and "nyangkut." Meanwhile, the Decision Tree classification model provides better performance.Keywords : investor, lexicon, social network, stock exchange, text miningCorrespondence to : aestikani.mahani-2019@feb.unair.ac.id Penurunan optimisme investor pasar modal adalah salah satu dampak keuangan pada dunia usaha yang timbul akibat pandemi SARS-COVID19. Hal ini tercermin dari turunnya volume perdagangan yang diikuti penurunan tajam IHSG di Bursa Efek Indonesia mulai Maret 2020. Sehingga kekhawatiran atas perlambatan pemulihan ekonomi sebagai dampak pandemi, tercermin dari sentimen investor di pasar modal. Di satu sisi, perkembangan internet di Indonesia yang pesat, memicu kecenderungan aktivitas investor dalam pencarian informasi sebelum membeli dan menjual surat berharga secara online, turut berkontribusi dalam mempengaruhi preferensi dan sentimen investor. Penelitian ini menggali ekspektasi investor yang tercermin pada sentimen investasi, dimana pasar modal sebagai salah satu barometer penting perekonomian suatu negara. Kajian ini mengeksplorasi fitur/terms investasi saham yang kerap muncul di pasar modal dan mengumpulkannya dalam kamus leksikon. Kemudian, dilakukan ekstraksi opini investor berbasis leksikon yang digali dari jejaring sosial Twitter, dilanjutkan dengan tahap text mining yaitu menganalisis sentimen, dan membentuk model klasifikasi berbasis Naive Bayes dan Decision Tree. Keluaran penelitian ini menunjukkan bahwa polaritas sentimen investor pasar modal adalah positif dengan fitur sentimen yang sering muncul yaitu “cuan”, “bearish”, “serok”, “copet”, “untung”, dan “cut loss”. Sedangkan model klasifikasi Decision Tree memberikan performansi akurasi yang kebih baik.Kata Kunci : Analisis sentimen; Investor; Leksikon; Text mining; Twitter

show abstract

“…[13] Penelitian Hong,dkk(2016) juga sejalan dengan Shi, Yong (2018) yang menghasilkan bahwa dengan menggunakan text mining, didapatkan sentimen investor memiliki efek positif pada jangka panjang dan sebaliknya pada jangka menengah . [12] Penelitian lain yang berjudul " A Text Mining Aplication on Monthly Price Development Report" oleh Hatice Burcu Eskici(2018) yang bertujuan menganalisis monthly price development report (MDPR) dari Central Bank of Republic Turki (CBRT) menyimpulkan bahwa kata atau group kata yang sering digunakan mampu merepresentasikan pengukuran inti dari inflasi serta beberapa sektor lain seperti perhotelan dan akomodasi. Selain itu, penggunaan trend, seasonality dan clustering MDPRs lebih signifikan dibandingkan yang hanya menggunakan trend dan seasonality saja untuk menggambarkan IHK tahunan.…”

Section: Pendahuluanunclassified

Sentiment Analysis Using Twitter Data Regarding BPJS Cost Increase and Its Effect on Health Sector Stock Prices

Wardhani¹,

Areka²,

Nugroho³

et al. 2020

IJAIDM

View full text Add to dashboard Cite

News about the increase in BPJS that will increase 2x gives a variety of responses in the community. One of the social media that people use in responding is Twitter. This research is used to see people's sentiment on Twitter about BPJS tariff policies. In addition, the impact of this sentiment will also be seen on the price of health shares. The analysis used is descriptive analysis and inference analysis. Descriptive analysis is used to look at the general picture of community sentiment and inference analysis is used to see the impact of community sentiment on the price of health stocks, namely Indo Farma and Kimia Farma. The results of this study indicate that public sentiment towards rising BPJS is dominated by negative sentiment. And for the level of tendency that has been processed through binary logistic regression analysis shows that negative sentiment will make Kimia Farma shares will go down while positive sentiment will make Kimia Farma shares will go up. As for IndoFarma shares, positive and negative sentiments from IndoFarma shares will tend to fall.

show abstract

A Text Mining Based Study of Investor Sentiment and Its Influence on Stock Returns

Abstract: Abstract. This paper studies a broad sample of investors' online opinion

Cited by 19 publications

References 7 publications

Sparse Principal Component Analysis for Natural Language Processing

Sparse Principal Component Analysis for Natural Language Processing

Prediksi Sentimen Investor Pasar Modal Di Jejaring Sosial Menggunakan Text Mining

Sentiment Analysis Using Twitter Data Regarding BPJS Cost Increase and Its Effect on Health Sector Stock Prices

Contact Info

Product

Resources

About