Technological Changes in Agriculture and Information Technology: Centrality and Citation Analyses of South Korean Agricultural Patent Data

Abstract

This study analyses recent changes in the agricultural technology trends at the current pace of its convergence with information technology (IT) using South Korean agricultural patent data from 2000 to 2015. This article focuses on the structural changes in agricultural technology in terms of patent applications by specific technology sector between the two periods: Period 1 (2000–2007) and Period 2 (2008–2015). Accordingly, we performed centrality analysis to measure the importance of each agricultural technology based on International Patent Classification (IPC) sub-classes. We also conducted citation analysis to identify whether the proportion of backward and forward citations of agricultural patents had significantly changed between the two periods. The results of the centrality analysis suggest that, whereas food technology was the most important technology sector in agriculture during both the periods, agricultural production technology experienced a considerable increase in its share in agriculture in Period 2. The results of the citation analysis confirm a substantial degree of interconnectivity between agricultural technology and non-agricultural technologies.

Keywords

Technological change agricultural patents centrality analysis citation analysis information technology

Introduction

The agricultural sector in South Korea (hereafter, Korea) has recently experienced considerable changes in terms of technology as it converges with information technology (IT). The phenomenon of agricultural technology convergence with non-agricultural sectors is gaining steady progress and one of the most active cases may be agriculture-IT convergence (e.g., spraying drones, smart farm).

In general, technological changes are recognised as changes in the relative importance of specific technologies within certain technology sectors, which have been investigated through a network analysis using patent or research and development data (Kim, Jong, Jang, & Lee, 2016b; Lee, Kim, & Jang, 2016; Li, Lin, Chen, & Roco, 2007). However, unlike technological changes in non-agricultural sectors such as manufacturing, agricultural technology changes have not drawn much attention in this line of research. On the one hand, the paucity of relevant literature on agricultural technology may be due to a low interest in the agricultural sector. On the other hand, there is a perception that technological changes in the agricultural sector are less influential than those occurring in non-agricultural sectors. However, following the emergence of new technologies brought about by digital transformation, technological changes in agriculture have recently drawn increasing attention in many countries and are expected to greatly influence agriculture in Korea (Lee, 2017; Lim, 2017).

The present article investigates recent structural changes in agricultural technology with a particular focus on its intra-sectoral and inter-sectoral technology connectivity. For the analysis of intra-sectoral technology change in agriculture, we analyse the changes in the importance of each agricultural technology within the agricultural sector using centrality analysis. For the analysis of inter-sectoral technology connectivity, we conduct citation analysis to examine how the degree of convergence between agricultural technology and non-agricultural technologies has changed. Furthermore, changes over time in patent citation were confirmed by Pearson’s chi-squared (χ²) test. With an emerging trend in the convergence of IT in agricultural technology and a growing number of South Korean (hereafter, Korean) farmers using smartphones and other mobile devices (e.g., tablet PC) (Statistics Korea, 2015), we separately examined the citation relationship between agricultural and IT patents.

The data used in the present study were Korean agricultural patent data applied from 2000 to 2015, which were further divided into Period 1 (2000–2007) and Period 2 (2008–2015) to examine changes over time. Agricultural patents are defined as five classes under Section A of the International Patent Classification (IPC),¹ including ‘agricultural production’ (A01), ‘baking or doughs’ (A21), ‘butchering or meat treatment’ (A22), ‘foods or foodstuff’ (A23) and ‘tobacco’ (A24).

The remainder of this article is organised as follows. The second section provides a general discussion on agricultural technology and presents some recent case examples of agriculture-IT convergence. The third section explains the data and methodology used in this study. The fourth section summarises the results of centrality analysis and citation analysis. In the fifth section, we draw conclusions.

Literature Review

Technological Advances and Patents in Agriculture

Agricultural patents have been widely used as a measure of agricultural technology, especially in the field of agricultural biotechnology. Unlike agricultural biotechnology patents, agricultural patents in more basic areas, such as crop production, had drawn little attention. Recently, more intensive analyses on both qualitative and quantitative agricultural patents have started to emerge.

Barham, Foltz, and Kim (2002) highlighted qualitative growth in agricultural patents by examining annual trends in university-granted agricultural biotechnology (ag-biotech) patents sorted by the US Patent and Trademark Office (USPTO). According to the results of this study, the frequency of application for ag-biotech patents applied by the universities has dramatically increased after 1996, resulting in a larger number of forward citations in ag-biotech patents as compared to that in other sectors. By contrast, Buccola and Xia (2004) showed that patent quality in ag-biotech declined as the quality is measured by the number of forward citations of ag-biotech patents. The authors argued that such a decline in quality can be attributed to both innate technological factors in the ag-biotech industry and firms’ strategic factors related to patent protection. Furthermore, in her analysis of the probability of patent application of nine multi-national bio-agricultural firms using a probit model, Chan (2010) found that the fixed effect of each country where a firm is located has a large impact on that firm’s patent applications. Meanwhile, Liu, Cao, and Song (2014) conducted a duration analysis to measure the quality of agricultural patents applied in China. The authors found that between 1985 and 2005, the number of agricultural patent applications in China increased by more than 30 times (from 146 to 5,510), while during the same period, a seventyfold increase (from 32 to 2,244) was observed in the number of ag-biotech patents. In addition, while the quality of agricultural patents measured by patent life span and length of patent renewal steadily increased, the quality of agricultural patents applied by non-nationals was higher than those applied by their own nationals.

Agricultural Technology and IT

The development of IT has brought the convergence of agricultural technology and IT, especially in the agricultural production sector (Kuhlmann & Brodersen, 2001; Ojha, Misra, & Raghuwanshi, 2015; Pham & Stack, 2018; Singh, Sankhwar, & Pandey, 2014; Wang, Zhang, & Wang, 2006). This phenomenon, the convergence between the two technologies, has been called precision agriculture technologies (PATs) in previous studies. Plant (2001) and Murakami et al. (2007) defined PATs as the management systems using IT to collect agricultural data for using in site-specific crop management. From a similar point of view, Paustian and Theuvsen (2017) identified PATs as the IT control systems managed by farmers for lowering the production cost and enhancing the profitability of crop sales.

The main reason why PATs are attracting attention is that the convergence between agricultural technology and IT builds an efficient communication system in the agricultural sector, which includes entire relationships between factors and products, products and farmers and farmers and farmers (Cox, 2002). Similarly, Huggins and Valverde (2018) proposed the complexities as an underlying factor for the increasing significance of IT in the agricultural sector. That is, the multiple goals of agriculture such as improving nutrition, food security and livelihood for rural women and children create complexities in the agricultural sector, which requires IT to manage itself. In addition, IT can be easily converged to agricultural technologies since agricultural productivity can grow with it. For example, representative monitoring systems, automatic temperature control systems and real-time distribution systems have already been evaluated as valuable technologies in view of enhancing agricultural productivity. In this regard, Jiayu, Shiwei, Zhemin, Wei, and Dongjie (2015) argued that the monitoring systems help production components such as material inputs, environment, climate and policy to have real-time feedback with each other, and thereby maximise the agricultural productivity.

In particular, the expansion of IT in the agricultural sector is naturally leading to the emergence of technologies related to digital transformation. Although there is a lack of empirical evidence for this trend, some studies have introduced the technologies through several practical examples (Krintz et al., 2016; TonKe, 2013; Wolfert, Ge, Verdouw, & Bogaardt, 2017; Yan-e, 2011). Yan-e (2011) introduced an automated system that identifies the information of crops and determines the appropriate amount of fertiliser input through the sensors on a tractor hopper. Considering software technology, TonKe (2013) insisted that large amounts of agricultural data could be obtained from the Internet of Things (IoT) and the cloud system embedded in the agricultural production sector. By combining the technologies of digital transformation, studies which highlight smart farm have been emerging in recent years. In a representative study, Krintz et al. (2016) defined smart farm with the following two conditions. First, the technologies with the convergence between agricultural technology and IT should be concentrated in one place. Second, farmers should be able to easily manipulate these technologies via applications on their individual IT devices. Furthermore, Wolfert et al. (2017) predicted that smart farm will be used in more diverse areas (not just production) such as in distribution due to the accumulation of big data in agriculture.

These technologies are also being used in agriculture in Korea. According to Kim, Park, and Park (2016c), as of 2015, information and communications technology (ICT)-based smart farms were operated on 2,625 horticulture farms. The adoption of smart farms increased the farms’ net income by 40.5 per cent and reduced monthly working days by 5.4 per cent. Big data technologies are also being applied to agriculture in Korea. For instance, the information of insect pests and price trends is being provided to farmers via Internet webpages (Kim, Koo, An, & Han, 2016a).

Data and Methodology

Data

In the present study, technological changes in agriculture were measured by agricultural patents applied from 2000 to 2015. Agricultural patents were defined as patents that correspond to five patent classes in Section A of the IPC code. Specifically, agricultural patents include ‘agricultural production’ (A01), ‘baking or doughs’ (A21), ‘butchering or meat treatment’ (A22), ‘foods or foodstuff’ (A23) and ‘tobacco’ (A24).² A total of 94,828 agricultural patents were extracted from these categories, in which 66,092 had backward citations and 49,323 had forward citations. Since each patent may include more than one citation (either backward or forward), the total number of observations used for the present analysis amounted to 267,580 backward citations and 174,155 forward citations.

Table 1 shows the number of agriculture patents applied for the years from 2000 to 2015 based on classes and sub-classes of IPC. At the class level of IPC, ‘agricultural production’ (A01) and ‘foods or foodstuff’ (A23) are the two dominant technology fields in agricultural technology, assuming 93.6 per cent of total patent applications in agriculture in the entire 15-year period under study. Furthermore, ‘agricultural production’ (A01) with 46,421 cases (49.0%) presented the largest percentage of the total number of agricultural patents for the observed period. This implies that production technology is the key technology in agriculture due to the substantial technology advancement since 2000. ‘Foods or foodstuff’ (A23), including 42,323 agricultural patents, accounted for another 44.6 per cent of the total number of agricultural patents, indicating that agricultural technology is also critical for the processing and commercialisation of agricultural products.

Table 1

Agricultural Patent Applications in Korea (Accumulated for 2000–2015)

IPC	Description	Number of Applications	Proportion (%)
A01	Agriculture; forestry; animal husbandry; hunting; trapping	46,421	49.0
A01B	Soil working in agriculture or forestry	1,709	1.8
A01C	Planting; sowing; fertilising	1,830	1.9
A01D	Harvesting; mowing	1,996	2.1
A01F	Threshing; baling of straw, hay or the like	734	0.8
A01G	Horticulture; cultivation; forestry; watering	15,251	16.1
A01H	New plants or processes for obtaining them	1,229	1.3
A01J	Manufacture of dairy products	74	0.1
A01K	Animal husbandry; care of animal; fishing	15,316	16.2
A01L	Shoeing of animals	6	0.0
A01M	Catching, trapping or scaring of animals	3,013	3.2
A01N	Preservation of bodies of humans or animals or plants	5,132	5.4
A01P	Biocidal, pest repellent	131	0.1
A21	Baking; equipment for making or processing doughs	2,369	2.5
A21B	Bakers’ ovens; machines or equipment for baking	410	0.4
A21C	Machines or equipment for making or processing doughs	640	0.7
A21D	Treatment of flour or doughs for baking	1,319	1.4
A22	Butchering; meat treatment; processing poultry or fish	1,208	1.3
A22B	Slaughtering	76	0.1
A22C	Processing meat, poultry, or fish	1,132	1.2
A23	Foods or foodstuffs; their treatment, not covered by other classes	42,323	44.6
A23B	Preserving; chemical ripening of fruit or vegetables	4,370	4.6
A23C	Dairy products; milk or cheese substitutes; making thereof	1,063	1.1
A23D	Edible oils or fats	536	0.6
A23F	Coffee; tea; their substitutes; manufacture, preparation, or infusion	1,826	1.9
A23G	Cocoa; cocoa products; confectionery; chewing gum	2,485	2.6
A23J	Protein compositions for foodstuffs	254	0.3
A23K	Feeding-stuffs specially adapted for animals	2,224	2.4
A23L	Foods, foodstuffs, or non-alcoholic beverages	26,917	28.4
A23N	Machines or apparatus for treating harvested fruit, vegetables	1,920	2.0
A23P	Shaping or working of foodstuffs	728	0.8
A24	Tobacco; cigars; cigarettes; smokers’ requisites	2,507	2.6
A24B	Manufacture or preparation of tobacco for smoking or chewing	384	0.4
A24C	Machines for making cigars or cigarettes	90	0.1
A24D	Cigars; cigarettes; tobacco smoke filters	861	0.9
A24F	Smokers’ requisites; match boxes	1,172	1.2
Total		94,828	100.0

Source: Conducted by the authors.

Note: Descriptions of the classification were abbreviated to the extent that it does not undermine its original meaning.

At the sub-class level of IPC, ‘cooking’ (A23L) accounted for the highest percentage of the total number of agriculture patents (28.4%). This category includes agricultural patents related to non-alcoholic beverages, modification of nutritive qualities and preservation of foods, all of which require advanced technologies for the secondary processing of agricultural products. Furthermore, ‘livestock’ (A01K) accounted for 16.2 per cent of the total, followed by 16.1 per cent for ‘cultivation or watering’ (A01G).

Figure 1 shows the number of agricultural patent applications per year from 2000 to 2015 by the type of applicants: individuals and organisations (government or firms).³ As shown in Figure 1, the total number of agricultural patent applications increased during the early 2000s, but then started to decrease in 2004. In subsequent years, it has steadily grown from 2007 and onwards, except for 2015. It is of interest to note that agricultural patents applied by individuals have continued to rise since 2000, exceeding the number of patents applied by organisations in 2012 and afterwards.

Figure 1

Source: Conducted by the authors.

Figure 2 illustrates the trends in the number of agricultural patent applications by IPC class, where the number of agricultural patent applications was converted into common logarithms. As shown in Figure 2, ‘agricultural production’ (A01) and ‘foods or foodstuff’ (A23), the two dominant technology fields, show stable trends, and technology fields like ‘baking or doughs’ (A21), ‘butchering or meat treatment’ (A22) and ‘tobacco’ (A24) exhibit greater variability.

Figure 2

Source: Conducted by the authors.

Methodology

This study examined the structural changes in agricultural technology by comparing the patent applications in Period 1 (2000–2007) and those in Period 2 (2008–2015). The reasons why the application years were divided into these two periods are that the number of agricultural patent applications turned to a stable upward trend from 2007 (Figure 1) and applications of personal information devices, such as smart phones, had rapidly spread in the starting of mid-2000s. Thus, it is highly likely that both quantitative and qualitative changes in Korean agricultural patents have occurred during this period.

For the analysis of the relative importance of specific technologies, centrality analysis is generally used. This method derives the relative importance of each technology by using vertex and edge in a specific network.⁴ To determine the relative importance of each node in a particular network, the indicators of the centrality analysis are divided into the centrality between all the nodes and the power representing the dominance over the neighbouring nodes (Bonacich, 1987; Cook, Emerson, Gillmore, & Yamagishi, 1983). These indicators have been extensively used to analyse a wide range of subjects, including gene regulatory networks (Koschützki & Schreiber, 2008), world city networks (Neal, 2011), sales networks of department stores (Cho & Bang, 2011) and patent networks (Lee et al., 2016). In particular, using the US patents, Lee et al. (2016) measured the relative importance of different technology groups to find that the importance of medical technologies has considerably increased in recent years.

In the present study, agricultural technologies based on IPC sub-classes became vertex, and citation information of each technology was defined as edge.⁵ We used recursive centrality and recursive power proposed by Neal (2011) to measure the importance of each technology in agriculture. These indices offer great advantages in that they consider the centrality of the connected vertices when calculating the centrality indices of certain vertices (Neal, 2011).

Specifically, recursive centrality of an agricultural technology is the sum of its networks after weighting the degree centrality of other connected agriculture technologies and is closely related to information diffusion for other agricultural technologies (Neal, 2011). The same logic applies to the recursive power, but the inverse of the degree centrality of other connected agricultural technology is set as weight. The higher power of a certain agricultural technology means that the technology possibly dominates over other technologies (Derudder, Timberlake, & Witlox, 2010; Friedmann, 1986; Neal, 2011). The formulae for the recursive centrality and recursive power are shown in Equations (1) and (2).

R C_{i} = Σ_{j} R_{i j} \times D C_{j} ​

(1)

R P_{i} = Σ_{j} \frac{R_{i j}}{D C_{j}} ​

(2)

where RC_i and RP_i represent recursive centrality and recursive power of the ith vertex, respectively, and R_ij represents the entity in row i, column j of the network matrix of the entire vertex.⁶ DC_j depicts the degree centrality of the jth vertex, which is a quantitative measure of the network merely summing the edges of the jth vertex. In the present study, degree centrality is also presented with recursive centrality and recursive power. In addition, if an individual patent is defined as a vertex, there can be no bidirectional edge, since an individual patent cannot be cited in a later-applied patent (Lee et al., 2016). Therefore, we defined agricultural technology as a vertex that corresponds to the IPC sub-class level of the agricultural patent.

In addition, we examined the changes in the degree of convergence with other non-agricultural technologies of agricultural technology. For this analysis, we performed citation analysis to identify whether the proportion of backward and forward citations for non-agricultural patents of agricultural patents had changed significantly over time. According to Karvonen and Kässi (2013), while the forward citation of patents evaluates the significance of industry transformation, the backward citation reflects real convergence in technologies. For example, in their study measuring the degree of convergence between IT and biotechnology (BT) using the patent data from USPTO, Geum, Kim, Lee, and Kim (2012) found an increase in convergence between the two technologies from 2008 to 2010.

In the present study, to highlight the growing connectivity between agricultural technology and IT, the IT sector was separated from other non-agricultural sectors in the citation analysis. IT patents refer to the patents related to information, communication, signal and electrical control, based on IPC classes or sub-classes.⁷ The change in the proportion of backward and forward citations of agricultural patents between Period 1 and Period 2 was confirmed by Pearson’s chi-squared (χ²) test.

Results

Through the agricultural citation network, we can observe organic relationships between each agricultural technology as represented by agricultural patents. Figure 3 shows the network for Period 1 and Period 2 based on the adjacency matrix of the backward citation network of the total agricultural network (IPC sub-classes). For example, if i technology in the agricultural network cites j technology, element a_ij of the adjacency matrix has the value of 1. Otherwise, element a_ij has the value of 0. According to Figure 3, the relationships between agricultural technologies strengthened in 2008–2015, as compared to 2000–2007. In addition, while food technologies such as A23L and A23B were concentrated in the centre of the network, production technologies such as A01G and A01L slightly moved towards the centre in 2008–2015.

Figure 3

Source: Conducted by the authors.

Table 2 lists the five most important agricultural technologies derived from the centrality analysis for each period.⁸ In both periods, the technology corresponding to ‘cooking’ (A23L) turned out to be the most important technology in agriculture, as reported in all three indices. This is likely due to the need for many complex technologies in preservation, storage and transportation of foods. Other than ‘cooking’ (A23L), the technology related to ‘livestock’ (A01K) was important in all indices for 2000–2007. However, the technology related to ‘cultivation or watering’ (A01G) became important in 2008–2015. Although the importance of food-related technology was still high in both periods, it is noteworthy that the importance of technology related to production increased. In addition, the increased importance of technology related to ‘pesticides or herbicides’ (A01N) in 2008–2015 supports that the importance of production technology in agriculture increased. On the other hand, it should be noted that the importance of the technology related to ‘livestock’ (A01K) has declined in all indices in 2008–2015.

Table 2

List of Top Five Important Agricultural Technologies: Results of the Centrality Analysis

Rank	2000–2007			2008–2015
Rank	Degree Centrality	Recursive Centrality	Recursive Power	Degree Centrality	Recursive Centrality	Recursive Power
1	A23L	A23L	A23L	A23L	A23L	A23L
2	A01K	A01K	A01K	A01G	A01G	A01N
3	A01G	A01G	A01N	A01K	A01K	A01G
4	A23B	A21D	A24D	A23B	A23B	A24F
5	A21D	A23B	A01G	A01M	A23G	A01K

Source: Conducted by the authors.

Note: A01G (cultivation or watering), A01K (livestock), A01M (catching or trapping of animals), A01N (pesticides or herbicides), A23B (preserving of foods), A23L (cooking), A23G (cocoa or cocoa products), A24D (cigars or cigarettes), A24F (smokers’ requisites).

Figure 4 illustrates the recursive power indices of all agricultural technologies.⁹ In the centre of the circle, the indices of recursive power are large, and the overall recursive power of the production technology was strengthened in Period 2 as compared with Period 1.

Figure 4

Source: Conducted by the authors.

The growing importance of agricultural production technology can be interpreted as internal and external factors. External factors are those new agricultural production technologies that emerge from agriculture-IT convergence since these new technologies are mainly emerging in the production sector. In particular, digital transformation reshapes the agricultural production technologies as it is evidenced from the emergence of unmanned tractors and spraying drones. Furthermore, past studies suggest that digital transformation can spur the flexibility and productivity growth by strengthening the organic relationship between workers and production systems (Gorecky, Schmitt, Loskyll, & Zühlke, 2014; Monostori, 2014; Shrouf, Ordieres, & Miragliotta, 2014; UBS, 2016). Internal factors are those involved with strengthening the agricultural production systems. In Korea, technology innovation in agriculture used to focus on food processing rather than farming practices in the past. However, the recent trend involves eco-friendly farming practices, precision agriculture and seed breeding techniques which requires high-level technical and managerial skills at the farm-level (Huggins & Valverde, 2018; Murakami et al., 2007; Paustian & Theuvsen, 2017).

Agricultural technology also undergoes external changes in that it becomes increasingly converged with other non-agricultural technologies. Table 3 shows the changes in the proportion of backward citation of agricultural patents for the agricultural, IT and non-IT sectors. As ‘agricultural production’ (A01) includes various sub-technologies with very different characteristics, it was divided into ‘agricultural production’ (A01_1) and ‘animal husbandry’ (A01_2).¹⁰

Table 3

Results of the Backward Citation Analysis of Agricultural Patents

			Agricultural Patents (%)	IT Patents (%)	Non-IT Patents (%)
Total	Period 1	88,507	76.6	2.5	20.9
	Period 2	179,073	77.0	2.6	20.4
	Pearson’s chi-squared (χ²)		6.2**	1.6	9.5***
A01_1 (production of agriculture)	Period 1	24,434	71.9	3.1	25.0
	Period 2	56,730	72.0	3.2	24.8
	Pearson’s chi-squared (χ²)		0.07	1.5	0.6
A01_2	Period 1	12,061	79.7	3.0	17.3
(animal husbandry)	Period 2	25,970	76.4	3.5	20.1
(animal husbandry)	Pearson’s chi-squared (χ²)		50.7***	5.8**	41.6***
A21 (banking or doughs for baking)	Period 1	2,049	80.1	2.8	17.1
	Period 2	4,703	83.7	2.8	13.5
	Pearson’s chi-squared (χ²)		12.7***	0.00	14.9***
A22	Period 1	970	78.9	3.9	17.2
(butchering or meat treatment)	Period 2	2,256	73.9	4.0	22.1
(butchering or meat treatment)	Pearson’s chi-squared (χ²)		8.9***	0.0	9.8***
A23	Period 1	47,467	77.8	2.1	20.1
(food or foodstuffs)	Period 2	86,619	80.1	1.9	18.0
(food or foodstuffs)	Pearson’s chi-squared (χ²)		97.8***	5.3**	89.3***
A24	Period 1	1,526	79.8	2.7	17.5
(tobacco)	Period 2	2,807	77.1	2.4	20.5
	Pearson’s chi-squared (χ²)		4.3**	0.3	5.6**

Source: Conducted by the authors.

Notes: Only agricultural patents with backward citations were included.

*** p<0.01; ** p<0.05; * p<0.1.

Over the two periods, the frequencies of backward citation of total agricultural patents more than doubled from 88,507 to 179,073.¹¹ While the number of agricultural patent applications increased over time, the frequencies of backward citation per agricultural patent also increased for all sectors in the same period. In other words, with the continuous accumulation of new technologies, the agricultural sector has become increasingly dependent upon relevant agricultural and non-agricultural technologies.

In the case of total agricultural patents, the proportion of backward citation for agricultural patents increased, while the corresponding proportion decreased for non-IT patents. However, the difference in the proportion of backward citation for IT patents was not statistically significant. This suggests that the convergence of agricultural technology is more noticeable within the agricultural sector. At the sub-class level of IPC, no change in the proportion of backward citation of ‘agricultural production’ (A01_1) for the aforementioned three sectors was observed. In the case of ‘foods or foodstuff’ (A23), the proportion of backward citations for the agricultural sector increased. However, it significantly decreased in the case of ‘animal husbandry’ (A01_2), ‘butchering or meat treatment’ (A22) and ‘tobacco’ (A24). Specifically, the proportion of citing IT patents of ‘animal husbandry’ (A01_2) significantly increased. In summary, the degree of convergence and its nature of agricultural technology vary depending on each subcategory of agricultural technology.

Table 4 presents the changes over time in the proportion of agricultural patents cited by agricultural, IT and non-IT patents. The frequencies of forward citation of the total number of agricultural patents were 121,070 in Period 1, decreasing afterwards to 53,085 in Period 2. The number of agricultural patent applications has continued to increase due to the decrease in forward citations per agricultural patent. However, the forward citation statistics tend to be underestimated, especially for recent ones and do not reflect sufficient time lag required for patent citation.

Table 4

Results of the Forward Citation Analysis of Agricultural Patents

			Agricultural Patent (%)	IT Patents (%)	Non-IT Patents (%)
Total	Period 1	121,070	82.5	1.5	15.9
	Period 2	53,085	76.7	4.1	19.3
	Pearson’s chi-squared (χ²)	817.6***	1.1e+03***	289.2***
A01_1	Period 1	33,106	79.1	1.7	19.3
(production of agriculture)	Period 2	16,564	73.3	4.9	21.8
(production of agriculture)	Pearson’s chi-squared (χ²)	205.8***	434.8***	42.7***
A01_2	Period 1	17,154	84.0	2.0	14.0
(animal husbandry)	Period 2	7,112	75.3	5.4	19.4
(animal husbandry)	Pearson’s chi-squared (χ²)	256.6***	197.6***	111.7***
A21	Period 1	2,816	88.4	1.5	10.1
(banking or doughs for baking)	Period 2	1,434	87.3	3.2	9.5
(banking or doughs for baking)	Pearson’s chi-squared (χ²)	1.1	14.5***	0.4
A22	Period 1	1,207	86.5	0.7	12.8
(butchering or meat treatment)	Period 2	483	73.3	4.4	22.4
(butchering or meat treatment)	Pearson’s chi-squared (χ²)	42.1***	27.8***	23.8***
A23	Period 1	65,430	83.7	1.3	15.0
(food or foodstuffs)	Period 2	26,327	78.8	3.3	18.0
(food or foodstuffs)	Pearson’s chi-squared (χ²)	311.5***	378.3***	125.7***
A24	Period 1	1,357	78.8	1.7	19.5
(tobacco)	Period 2	1,165	74.5	3.1	22.4
	Pearson’s chi-squared (χ²)	6.4**	5.3**	3.1*

Source: Conducted by the authors.

Notes: Only agricultural patents with forward citations were included.

*** p<0.01; ** p<0.05; * p<0.1.

Given this constraint, according to the results of Pearson’s chi-squared (χ²) test, the proportion of forward citations of total agricultural patents for the agricultural sector decreased, while that for the IT sector and non-IT sectors has significantly increased. In particular, the proportion of forward citation of total agricultural patents for the IT sector has shown a threefold increase from 1.52 per cent in Period 1 to 4.07 per cent in Period 2. This underscores the role of agricultural technology in convergence with other technologies, especially IT. On the class level of IPC, the proportion of forward citations for the agricultural sector decreased in all agricultural categories except for ‘baking and dough’ (A21), and the proportion for the IT sector and non-IT sectors significantly increased.

In summary, while food technologies continue to occupy the most important position in the agricultural technology network, production technologies show an increase of importance. As evidenced by the trend of backward citation of agricultural patents, agricultural technology has become increasingly reliant on related technologies, in both the agricultural sector and non-agricultural sectors.¹² However, the degree and nature of such interdependence vary by the subcategories of agricultural technology. Finally, in terms of forward citation, agricultural technology has also shown to converge with non-agricultural sector technologies such as IT.

Conclusions

This study investigated recent changes in the trends of agricultural technology in Korea in terms of both intra- and inter-sectoral structure changes. The data used in the present study were the Korean agricultural patent data applied for the years from 2000 to 2015, which were further divided into Period 1 (2000–2007) and Period 2 (2008–2015). To identify intra-sectoral structural changes within the agricultural sector, we performed centrality analysis of an agricultural citation network. To investigate inter-sectoral structure change of the agricultural technology, we performed citation analysis. Thus, we regrouped the patent sectors into three sectors (agricultural, IT and non-IT), and we investigated sectoral changes in terms of the proportion of backward and forward citation of agricultural patents.

Key findings and implications of the present study are as follows. First, while the number of agricultural patent applications has shown a steady growth since the mid-2000s (except for 2015), the citation network among agricultural patents has expanded. This means that agricultural technology is rapidly developing and organic linkages within the agricultural sector are increasing. Second, according to the results of the centrality analysis, the importance of production technology (i.e., cultivation/watering, A01G) is substantially increasing, although food technology (i.e., foods/foodstuff, A23L) still occupies the largest share in agricultural technologies. The increasing importance of production technology within agriculture may reflect the impact of the convergence between agricultural technology and IT, especially digital transformation. In addition, the expansion of the precision agriculture technology in the production sector is also considered to be the main reason of convergence. Third, the citation analysis indicates that agricultural technology has increasingly converged with non-agricultural technologies such as IT. In the case of agricultural patents applied during 2008–2015, one-fourth of backward and forward citations were from non-agricultural technology sectors. The convergence between agricultural technology and IT accelerated particularly noticeably in the last several years.

The agricultural sector is not only rapidly accumulating new technologies, but is also experiencing structural changes. For instance, in Korea, the convergence between agricultural technology and IT such as automatic temperature control systems and real-time distribution systems has already been observed in the agricultural sector for several years. Especially, digital transformation technologies such as big data and artificial intelligence have already been adopted in agricultural production, which is the fastest growing sector in agricultural technology. Alongside the increasing connectivity of different technology fields within the agricultural sector, agricultural technology is increasingly converging with non-agricultural technology sectors. Under the influence of the IT and digital transformation, the bi-directional connectivity between agricultural and non-agricultural technologies is likely to continue strengthening.

As compared to the manufacturing and service sectors, the agricultural sector in Korea has steadily dwindled in its share of gross domestic product (GDP) and suffered from relatively low productivity. Therefore, the agricultural sector has been strongly criticised for impeding the continuing growth of Korea’s economy. However, the innovation capabilities in the agricultural sector for convergence to non-agricultural technologies and for adoption of digital transformation technology are expected to be a turning point for the sustainable growth of agriculture in Korea, causing unprecedented changes in agricultural technology.

This study is confined to agricultural patent data due to the data availability. Therefore, further research should be extended to an analysis of the patents in the entire industrial sector to further clarify the extent and direction of interconnectivity between agricultural and non-agricultural sectors. Also, if more patent data are accumulated in the future, a more detailed analysis of IPC group-based agricultural patents would be desirable. Such an analysis can reasonably be expected to draw practical implications for on-site applications of agricultural technologies.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

Appendix

Appendix D

Number of Forward Citations per Agricultural Patent by Sectors

		Agricultural Patents	IT Patents	Non-IT Patents
Total	Period 1	2.32	0.04	0.45
	Period 2	0.79	0.04	0.20
A01 (agricultural production)	Period 1	1.86	0.04	0.40
	Period 2	0.71	0.05	0.20
A01_1 (agricultural production)	Period 1	1.91	0.04	0.47
	Period 2	0.70	0.05	0.21
A01_2 (animal husbandry)	Period 1	1.78	0.04	0.30
	Period 2	0.74	0.05	0.19
A21 (baking or doughs)	Period 1	2.38	0.04	0.27
	Period 2	0.95	0.03	0.10
A22 (butchering or meat treatment)	Period 1	1.69	1.94	0.25
	Period 2	0.60	0.78	0.18
A23 (foods or foodstuffs)	Period 1	2.95	0.00	0.53
	Period 2	0.87	0.00	0.20
A24 (tobacco)	Period 1	1.03	0.84	0.26
	Period 2	0.59	0.59	0.18

Source: Conducted by the authors.

Notes

References

Barham

, Foltz

, & Kim

(2002). Trends in university Ag-biotech patent production. Review of Agricultural Economics, 24(2), 294–308.

Bonacich

(1987). Power and centrality: A family of measures. American Journal of Sociology, 92(5), 1170–1182.

Buccola

, & Xia

(2004). The rate of progress in agricultural biotechnology. Review of Agricultural Economics, 26(1), 37–44.

Chan

H. P.

(2010). The determinants of international patenting for nine agricultural biotechnology firms. Journal of Industrial Economics, 58(2), 247–278.

Cho

, & Bang

(2011). Applying centrality analysis to solve the cold-start and sparsity problems in collaborative filtering. Journal of Intelligent Information System, 17(3), 99–114.

Cook

K. S.

, Emerson

R. M.

, Gillmore

M. R.

, & Yamagishi

(1983). The distribution of power in exchange networks: Theory and experimental results. American Journal of Sociology, 89(2), 275–305.

Cox

(2002) Information technology: The global key to precision agriculture and sustainability. Computers and Electronics in Agriculture, 36(2–3), 93–111.

Derudder

, Timberlake

, & Witlox

(2010). Introduction: Mapping changes in urban system. Urban Studies, 47(9), 1835–1841.

Friedmann

(1986). The world city hypothesis. Development and Change, 17(1), 69–83.

10.

Geum

, Kim

, Lee

, & Kim

M. S.

(2012). Technological convergence of IT and BT: Evidence from patent analysis. ETRI Journal, 34(3), 439–449.

11.

Gorecky

, Schmitt

, Loskyll

, & Zühlke

(2014). Human-machine-interaction in the industry 4.0 era. 2014 IEEE International Conference on Industrial Informatics (INDIN). Porto Alegre: Institute of Electrical and Electronics, Engineers (IEEE).

12.

Huggins

, & Valverde

(2018). Information technology approaches to agriculture and nutrition in the developing world: A systems theory analysis of the mNutrition program in Malawi. Food Security, 10(1), 151–168.

13.

Jiayu

, Shiwei

, Zhemin

, Wei

, & Dongjie

(2015). Application of intelligence information fusion technology in agriculture monitoring and early-warning research. 2015 IEEE International Conference on Control, Automation and Robotics. Singapore: Institute of Electrical and Electronics, Engineers (IEEE).

14.

Karvonen

, & Kässi

(2013). Patent citations as a tool for analyzing the early stages of convergence. Technological Forecasting & Social Change, 80(6), 1094–1107.

15.

Kim

, Koo

, An

, & Han

(2016a). Big data applications in agricultural sector (KREI Research Report No.R802). Jeollanam-do: Korea Rural Economic Institute.

16.

Kim

, Jong

S. J.

, Jang

, & Lee

(2016b). Innovation of technology and social changes-quantitative analysis based on patent big data. Korean Journal of Applied Statistics, 29(6), 1025–1039.

17.

Kim

, Park

, & Park

(2016c). An analysis of the current status and success factors of smart farms (KREI Research Report No.M141). Jeollanam-do: Korea Rural Economic Institute.

18.

Koschützki

, & Schreiber

(2008). Centrality analysis methods for biological networks and their application to gene regulatory networks. Gene Regulation and Systems Biology, 2(1), 193–201.

19.

Krintz

, Wolski

, Golubovic

, Lampel

, Kulkarni

, Sethuramasamyraja

, … Liu

(2016). Smartfarm: Improving agriculture sustainability using modern information technology. 2016 KDD Workshop on Data Science for Food, Energy, and Water. San Francisco: Association for Computing Machinery (ACM).

20.

Kuhlmann

, & Brodersen

(2001). Information technology and farm management: Developments and perspectives. Computers and Electronics in Agriculture, 30(1–3), 71–83.

21.

Lee

(2017). The fourth industrial revolution and future of agriculture (Vol. 200, pp. 3–16). Jeollanam-do: World Agriculture, Korea Rural Economic Institute.

22.

Lee

, Kim

, & Jang

(2016). Patent citation network analysis. Korean Journal of Applied Statistics, 29(4), 613–625.

23.

, Lin

, Chen

, & Roco

M.C.

(2007). Worldwide nanotechnology development: A comparative study of USPTO, EPO, and JPO, patents (1976–2004). Journal of Nanoparticle Research, 9(6), 977–1002.

24.

Lim

Y. H.

(2017). Policy trends of agricultural science and technology in major countries in response to the fourth industrial revolution (Vol. 201, pp. 57–70). Jeollanam-do: World Agriculture, Korea Rural Economic Institute.

25.

Liu

, Cao

, & Song

(2014). China’s agricultural patents: How has their value changed amid recent patent boom? Technological Forecasting & Social Change, 88, 106–121.

26.

Monostori

(2014). Cyber-physical production systems: Roots, expectations and R&D challenges. Procedia CIRP, 17, 9–13.

27.

Murakami

, Saraiva

A. M.

, Ribeiro Junior

L. C. M.

, Cugnasca

C. E.

, Hirakawa

A. R.

, & Correa

L. P.

(2007). An infrastructure for the development of distributed service-oriented information systems for precision agriculture. Computers and Electronics in Agriculture, 58(1), 37–48.

28.

Neal

(2011). Differentiating centrality and power in the world city network. Urban Studies, 48(13), 2733–2748.

29.

Ojha

, Misra

, & Raghuwanshi

N. S.

(2015). Wireless sensor networks for agriculture: The state-of-the-art in practice and future challenges. Computers and Electronics in Agriculture, 118, 66–84.

30.

Paustian

, & Theuvsen

(2017). Adoption of precision agriculture technologies by German crop farmers. Precision Agriculture, 18(5), 701–716.

31.

Pham

, & Stack

(2018). How data analytics is transforming agriculture. Business Horizons, 61(1), 125–133.

32.

Plant

R. E.

(2001). Site-specific management: The application of information technology to crop production. Computers and Electronics in Agriculture, 30(1), 9–29.

33.

Shrouf

, Ordieres

, & Miragliotta

(2014). Smart factories in industry 4.0: A review of the concept and of energy management approached in production based on the internet of things paradigm. 2014 IEEE International Conference on Industrial Engineering and Engineering Management. Bandar Sunway: Institute of Electrical and Electronics, Engineers (IEEE).

34.

Singh

, Sankhwar

, & Pandey

(2014). The role of information communication technology (ICT) in agriculture. Global Journal of Multidisciplinary Studies, 3(4), 225–230.

35.

Statistics Korea (2015). Census of agriculture, forestry and fishery 2015. Daejeon: Statistics Korea.

36.

TonKe

(2013). Smart agriculture based on cloud computing and IoT. Journal of Convergence Information Technology, 8(2), 210–216.

37.

UBS (2016). Extreme automation and connectivity: The global, regional, and investment implications of the fourth industrial revolution. UBS White Paper for the World Economic Forum Annual Meeting 2016. Zurich: Union Bank of Switzerland (UBS).

38.

Wang

, Zhang

, & Wang

(2006). Wireless sensors in agriculture and food industry-recent development and future perspective. Computers and Electronics in Agriculture, 50(1), 1–14.

39.

WIPO (2017). International patent classification database 2017 version. Geneva: World Intellectual Property Organization (WIPO).

40.

Wolfert

, Ge

, Verdouw

, & Bogaardt

M. J.

(2017). Big data in smart farming-a review. Agricultural Systems, 153, 69–80.

41.

Yan-e

(2011). Design of intelligent agriculture management information system based on IoT. In 2011 Fourth International Conference on Intelligent Computation Technology and Automation. Shenzhen: Institute of Electrical and Electronics, Engineers (IEEE).