1. Introduction

Tomatoes are the most consumed vegetable worldwide, followed by onions. Onions are used as an essential ingredient in various dishes. Recently, prices of vegetables such as onions have skyrocketed in several countries around the world, and inflation has worsened, threatening global supply ^[1]. Onion price fluctuations are influenced by a variety of factors, both internal and external. A report published by the Korea Institute of Rural Economics ^[2] presented the factors for onion price fluctuations in terms of supply and demand. On the supply side, there was an increase in labor costs due to a shortage of agricultural workers, bad weather (rainy season, typhoons, cold snaps in winter, etc.), and a temporary reduction in cultivated area. On the demand side, changes in eating habits due to infectious diseases such as COVID-19 (increased consumption of HMR food and home meals), increased disposable income due to government subsidies, and prices of imported onions were seen. In fact, according to a Korean news article ^[3] reported in the first half of the previous year, onion storage has remained the same as demand has declined due to the sluggish restaurant industry due to COVID-19. When onion prices fell, reports emerged that farmers had plowed their onion fields just before harvest, and the article foreshadowed a political struggle going to Seoul (the government). In addition, according to a press release from the National Statistics Office ^[4], crop yields have been sluggish due to bad weather, resulting in a decrease in fresh weight per onion. The decline in onion prices was cited as production (−24.2%) and cultivated area (−4.3%) were lower than in 2021. As such, various issues related to the production and consumption of agricultural products have been reported, and these goods have the potential to affect the selling price in the consumer market. The world's onion production was 93,226,400 tons, with China producing the most (23,907,509 tons), but the yield per ha was much higher in Korea (1,65 kg) than in China (277,22 kg) ^[4]. In addition, among the vegetables produced in Korea, onions ranked ^[5] with the highest food supply per 5 people ^[6]. Onions are an important ingredient in Korean cuisine. Since onions are a price-sensitive item, the government is scrutinizing the intensity of fluctuations in consumer consumption-buying sentiment depending on the issues at the time.^{[1][2][3][4][5][6]}

2. The relationship between online news articles and the influence of consumer selling prices of agricultural products

The influence of mass media has been studied for a considerable amount of time. Blood and Phillips ^[7] found that economic news can potentially affect consumer sentiment and affect economic conditions in the long run. Gunther and Storey's 'Estimation Media Impact Model' (PMIM) may be relevant ^[8]. In the absence of direct information to measure the effectiveness of media, researchers have found that people generally tend to change their attitudes or behaviors by viewing media as having a significant impact on public opinion. Tal-Or et al. ^[9] used PMIM in the food sector to indirectly manipulate perceptions of the impact of news articles on expected sugar shortages. As a result, they manipulated perceived exposure to news stories to measure behavioral intent. Attavanich et al. ^[10] analyzed the market impact of media coverage related to swine flu on the future prices of pork, live cattle, corn and soybeans. As a result, media coverage had a temporary negative impact on pig futures prices, but little on other futures prices. Han Mu Moung Cho et al. ^[11] analyzed the correlation between changes in news keywords for avian influenza and changes in egg prices. As a result of analyzing news keywords based on the frequency of occurrence, the top keywords for avian influenza-related news were 'egg', 'price', and 'government'. When the keyword appeared, it turned out that the price of eggs skyrocketed. In addition, as a result of analyzing the relationship between news keywords and egg prices, it was found that keywords representing specific events, such as "Made in the United States", "holidays", and "prices", rather than keywords related to avian influenza, affect price flu. That is, the main problem, rather than the frequency of appearance, affected the market price.^{[7][8][9][10][11]}
Based on media influence theory, Walker ^[12] found that word-of-mouth is the final mechanism for opinion change, but the media drives discourse. He studied the relationship between the news media and skyrocketing house prices in the UK. An analysis of more than 30 news stories about the UK housing market between 000 and 1993,2008 found that the media caused real house price fluctuations, which influenced the media's opinion of the housing market. Park Jae-so et al. ^[13] extracted positive and negative sentiment indicators related to apartment sales in Seoul, South Korea, using a topic modeling technique of online newspaper articles. Through correlation analysis and causal analysis, the influence relationship between the positive and negative sentiment index and the apartment sales price was analyzed. As a result of the analysis, it was found that the positive sentiment index has a significant effect on the sale price of small apartments in Seoul, excluding the city center. Seo Jeong-seok et al. ^[14] developed a positive/negative sentiment dictionary for article titles based on the combination of words that often appear in newspaper articles related to the housing market, and analyzed the dynamic relationship between housing market news and the housing market, and prepared a media report tone index for this purpose. Based on this, the relationship between real estate consumption sentiment, real transaction price index, and apartment transaction volume was analyzed. As a result of the analysis, the tone of media coverage of the housing market has a positive effect on apartment prices. In addition, Atri et al. ^[15] analyzed the impact of COVID-19 news, panic, and media coverage on oil and gold prices. The results showed that COVID-19 deaths and panic had a negative impact on crude oil prices.^{[12][13][14][15]}

Comprehensively considering the above previous research, it can be seen that media coverage can affect the market price. Recently, studies analyzing the relationship between media reports and market prices have been carried out by combining unstructured and structured data to complement traditional analytical methods and expand the scope of research. Therefore, in the field of agriculture, there was also a need for research on various aspects using big data.