Designing and Evaluating a Collaborative Knowledge Management Framework for Leaf Disease Detection

Bashir, Komal; Rehman, Mariam; Bashir, Afnan; Kanwal, Faria

doi:10.32604/csse.2022.022247

1Department of Computer Science, Lahore College for Women University, Lahore, 54000, Pakistan
2Department of Information Technology, Government College, Faisalabad, 54000, Pakistan
3Engage Research Lab, School of Law and Society, University of the Sunshine Coast, Sippy Downs, QLD 4556, Australia
*Corresponding Author: Komal Bashir. Email: komal.bashir@lcwu.edu.pk
Received: 01 August 2021; Accepted: 06 September 2021

Abstract: Knowledge Management (KM) has become a dynamic concept for inquiry in research. The management of knowledge from multiple sources requires a systematic approach that can facilitate capturing all important aspects related to a particular discipline, several KM frameworks have been designed to serve this purpose. This research aims to propose a Collaborative Knowledge Management (CKM) Framework that bridges gaps and overcomes weaknesses in existing frameworks. The paper also validates the framework by evaluating its effectiveness for the agriculture sector of Pakistan. A software LCWU aKMS was developed which serves as a practical implementation of the concepts behind the proposed CKMF framework. LCWU aKMS served as an effective system for rice leaf disease detection and identification. It aimed to enhance CKM through knowledge sharing, lessons learned, feedback on problem resolutions, help from co-workers, collaboration, and helping communities. Data were collected from 300 rice crop farmers by questionnaires based on hypotheses. Jennex Olfman model was used to estimate the effectiveness of CKMF. Various tests were performed including frequency measures of variables, Cronbach’s alpha reliability, and Pearson’s correlation. The research provided a KMS depicting KM and collaborative features. The disease detection module was evaluated using the precision and recall method and found to be 94.16% accurate. The system could replace the work of extension agents, making it a cost and time-effective initiative for farmer betterment.

Keywords: Collaborative knowledge management; framework; jennex olfman km success model; knowledge management; rice disease detection

Knowledge Management (KM) is a multidisciplinary area that involves individuals, knowledge communities, and organizations in the creation and management of knowledge of a particular discipline [1]. Advancements in technology and competitiveness in global economies have made the creation and diffusion of knowledge a critical component to consider while developing work practices [2]. Therefore, in today’s world knowledge could be considered as a commodity to design and develop new solutions, address existing problems, and bring innovation in existing processes. The transformation of knowledge involves various sources including individuals, relevant communities, research organizations, and companies responsible for production. The management of knowledge from multiple sources requires a systematic approach that can facilitate capturing all important aspects related to a particular discipline [3–5]. In this regard, researchers have already made advancements in creating KM frameworks that are facilitating various disciplines in improving their work practices, refining their products and services, and contributing to the economy in more effective and efficient manners [6–9].

Knowledge alone cannot assure the success of an organization; it is necessary to manage and share the knowledge [10]. The growth or even survival of any company depends on how efficiently it copes with the knowledge and information residing both inside and outside the organization [11]. Knowledge is generally considered as the most imperative component of competitive advantage for an economy [12,13]. Automating knowledge so that it can be shared with others is an important task [14]. Knowledge sharing is a process of knowledge exchange between individuals to create collective knowledge [15] is considered a key action in the process of KM. Through knowledge sharing, individuals can add their knowledge to knowledge repositories, enrich it with ideas, and ultimately add a competitive advantage for organizational growth [16]. Researchers have concluded that using a knowledge sharing strategy at multiple levels in an organization results reduction in production costs, speedy achievement of goals, enhanced team efficiency, an increase in revenue, and an increase in organizational growth [16–18]. Created and stored knowledge must be shared and disseminated to attain its maximum benefit [19].

Adding a communicative nature to KM gives a flavour of Collaborative Knowledge Management (CKM). Both tacit and explicit information can be shared collaboratively [20,21]. Modern business practices are collaborative in nature; the interaction between partners may be aimed at production or development, or it may simply be targeted at sharing resources or support services [22,23]. Organizations that adopt collaborative technologies must ensure compatibility between different systems and their priorities [21], and managers must find ways to manage collective intellectual knowledge assets [24]. KM is considered an important process for organizations and should be given prime consideration [18].

A KM framework is a system involving people, processes, and technology, as well as controls for effective KM implementation and execution [25]. It is responsible for the smooth flow of knowledge and the identification of gaps and interconnections among different elements [26]. The challenge for knowledge-based domains is to find an approach for the management and support of benefits based on the sharing of information in existing unmanaged activities in collaborative environments.

The focus of this research work is to develop a general framework termed CKMF (Collaborative Knowledge Management Framework) that minimizes the gaps identified in the existing KM frameworks. The primary goal is to facilitate collaboration among users as well as implement the framework using crop disease detection as a case study.

The rest of the paper is structured as follows: In Section 2, existing KM frameworks and their limitations are briefly discussed. In Section 3 materials and methods adopted for developing the framework are stated along with the discussion of the case study (used for the validation of the proposed concept) and hypotheses. In Section 4, the proposed framework is discussed and Section 5 holds results. Section 6 presents the discussion on results and finally, Section 7 provides the conclusion of this research.

A large amount of research has been conducted on developing KM systems to automate the KM and sharing processes. A number of frameworks have been designed by researchers to automate the KM concept, using processes and technologies that support knowledge asset management [27]. These frameworks are similar in many aspects, but they also consist of unique features and limitations [28]. Limitations in the frameworks lead towards need of improved framework. Tab. 1 presents the summarized view highlighting various features of studied KM frameworks.

The importance of KM becomes more significant in sectors where new tasks and projects are quite similar to the previous ones. In such cases, the lessons learned from past experiences are utilized to augment efficiency and effectiveness in future projects. It can significantly benefit the members of a community, as reusing the organizational knowledge can suggestively reduce both time and error, increasing the overall quality of the work that gives rise to the need for a collaborative system [21].

Considering the repetitive nature of agricultural activities and the significance of past trends and experiences for future farming decisions [29], CKMF (proposed in this research) is deployed in the agriculture sector of Pakistan as a case study to measure its effectiveness. Another important feature of the agriculture sector of Pakistan that necessitates KM infrastructure is lack of documentation, low literacy, and outdated extension agent network. A KM software, LCWU aKMS, was developed based on customized CKMF to cater needs of the farming community of Pakistan. A well-structured questionnaire instrument was used to collect feedback from rice crop farmers about the developed system. A sample of 300 rice growers was randomly selected to collect the desired data.

The research was carried out by adopting a systematic process-based approach. The research process designed is provided in Fig. 1. The proposed CKMF framework is constructed by studying various KM models and integrating their best features with new ones to overcome their issues. This was further refined by integrating knowledge from the agricultural discipline. A proof-of-concept implementation is carried out to realize the proposed framework. A case study was constructed around the agriculture sector in Pakistan. To evaluate the proposed framework, Information System (IS) model proposed by D&M was considered. It is required to determine the successfulness of a KM system to obtain maximum benefit from it [30]. The constructs of the framework were mapped with framework features to develop the proposed evaluation framework and construct relevant hypotheses.

A field survey was carried out to conduct a walkthrough and collect responses from framers associated with rice production farms. To collect data questionnaire was designed based on evaluation framework constructs. The questionnaires based on hypotheses were completed by 300 rice crop farmers to estimate the effectiveness of the designed software. Using the SPSS tool, quantitative data analysis is performed on the responses collected. The results section reports the details of various tests that were performed, including frequency measures of variables, Cronbach’s alpha reliability, and Pearson’s correlation.

A 5-point Likert scale is used to measure the latent constructs. All questionnaire items are measured based on the farmers’ level of agreement or disagreement, with a 5-point Likert scale ranging from (1-strongly agree to 5-strongly disagree). The measures for the survey constructs are obtained from prior studies, and the questionnaire consisted of two parts. The first part used nominal scales to collect the respondents’ demographic information, while the second part consisted of a subjective measure to evaluate the respondents’ perception regarding their evaluation of the LCWU aKMS as a KM system. A detailed account of the survey instrument is presented in Appendix A.

b) Develop an effective KM framework (CKMF) to overcome the limitations of existing KM frameworks

c) Proof the effectiveness of CKMF by making its customized implementation in the agriculture sectori)

i)Develop a CKMF based software (LCWU aKMS)

ii) Get user (farmer) feedback for LCWU aKMS

iii) Perform hypotheses-based evaluation based on KM success model (Jennex Olfman model) and statistical methods

iv) Check the accuracy of task performed by LCWU aKMS (accuracy comparison of rice leaf disease identification module)

The rest of this section explains the case study used in this research and the hypotheses-based evaluation mechanism.

Agriculture is the largest contributor to Pakistan’s economy; most of the population is dependent on this sector [31]. It accounts for approximately 24% of the total GDP and earns the largest portion of foreign exchange for Pakistan. Moreover, a major portion of the workforce is engaged in this sector [32].

ICT can benefit the agricultural sector greatly by helping advance research and assisting agriculture stakeholders in increasing production and supporting environmentally friendly methods. The main challenge faced by the agriculture sector is the lack of good methods for sharing knowledge. Many research activities are replicated due to the lack of an appropriate sharing mechanism [33]. On the other hand, farmers and field workers face problems when communicating their issues and knowledge with each other and the management. Moreover, extension workers, who act as intermediaries between farmers and authorities, further delay the response time. Decision-making for farmers depends upon their tacit knowledge [34]. The second challenge is inefficient procedures for sharing documents and guidance material. Although national agricultural bodies produce documents about the latest techniques for farmers, these documents are not properly distributed and managed [33]. A third challenge is communicating experiences and learning with other farmers and those new to farming, as there is no proper procedure to share and preserve one’s learning. Agriculture knowledge is mostly tacit; like knowledge of other domains, best practices are shared without documenting them [35].

Introducing KM technology in agriculture can help manage agriculture knowledge-sharing problems and enhance food production. The timely provision of information can reduce risk and uncertainty for farmers. To engage KM in agriculture, systematic mechanisms need to be developed to generate, capture, and share information [36–38]. The existing agriculture system in Pakistan was studied and it was found that it is difficult for farmers to get help promptly. The information flow for the existing system is presented in Fig. 2.

A research conducted for Punjab-based farmers by CABI South Asia states that the farmers consider themselves to be information and technology deprived [38]. They lack news about new technologies and crop-specific information also had negative views about the use of TV, radio, and newspaper mediums to access information [38]. Moreover, the issues faced by the extension agents include inappropriate incentives, high transportation costs, and insufficient training in technology and communication skills [38–40]. The research indicated highly positive farmer responses to the idea of using mobile phones to transmit information to farmers and described it as a low-cost solution [38].

The extension agent system has had little effect on crop production and has failed to develop effective research relationships [31]. Furthermore, extension services have been unable to successfully enhance farmers’ technological and technical expertise [41]. Over time, this sector has developed many problems and criticized for its unsatisfactory performance [42].

An evaluation model is required to estimate KM success. KM systems are considered a type of information system and the Jennex Olfman KM success. The KM success model is an extension of DeLone and McLean’s IS success model, which has a reputation as the benchmark for the evaluation of information systems [43–45]. The relationships between the constructs of the Jennex Olfman KM success model were used to develop the hypotheses to evaluate CKMF. The evaluation model presented in Fig. 3 is designed to validate and assess LCWU aKMS.

LCWU aKMS was implemented for the test case and responses were subsequently gathered from farmers using the system. These responses were collected using a questionnaire based on the designed hypotheses.

Defines how well the system performs the KM activities of knowledge creation, storage, retrieval, sharing, and application [46]. Technological resources include the ability of the organization to develop, operate, and maintain KM [43,46,47]. To determine the system quality of the LCWU aKMS, it is hypothesized that:

H1–There is a positive relationship between system quality and intent to use LCWU aKMS.

H2–There is a positive relationship between system quality and user satisfaction with LCWU aKMS.

Responsible for ensuring the right knowledge with sufficient context is gathered and is available for the right users at the right time [46,47]. For LCWU aKMS proposed hypotheses for knowledge quality are:

H3–There is a positive relationship between knowledge quality and intent to use LCWU aKMS.

H4–There is a positive relationship between knowledge quality and user satisfaction with LCWU aKMS.

Ensures that KM provides appropriate user support to effectively benefit from KM [47]. The following hypotheses are developed for service quality:

H5–There is a positive relationship between service quality and intent to use LCWU aKMS.

H6–There is a positive relationship between service quality and user satisfaction with LCWU aKMS.

This is a measure to obtain users’ perceptions of the benefits of KM, which help predict continued KM use. It helps measure the relationships between social factors involving knowledge use, perceived KM complexity, and benefits of knowledge use [47]. LCWU aKMS is evaluated for this measure using the following hypotheses:

H7–There is a positive relationship between intent to use and user satisfaction with LCWU aKMS.

H8–There is a positive relationship between intent to use and net benefits of LCWU aKMS.

User satisfaction dimension is used to measure users’ fulfilment with the KM system. It helps to measure KM use, as a desire to use KM depends on users being satisfied with KM [47]. The hypotheses for the user satisfaction dimension are:

H9–There is a positive relationship between user satisfaction and net benefits of LCWU aKMS.

H10–There is a positive relationship between system quality, knowledge quality, service quality, and intent to use LCWU aKMS.

H11–There is a positive relationship between system quality, knowledge quality, service quality, and user satisfaction with LCWU aKMS.

H13–There is a positive relationship between user satisfaction and intent to use LCWU aKMS.

The net benefits are associated with the use of knowledge may have good or bad benefits; feedback from these help organizations improve their KMS. Using the KM system has an impact on the individual and hence on the organization, therefore it is important to measure it and get feedback [47]. Success impact is measured for four KM areas including leadership, KM strategy, KM content, and process impact. This dimension provides feedback to the respective dimension so that adjustments can be made [47]. The hypotheses for net benefits assessment of LCWU aKMS are as follows:

H12–There is a positive relationship between intent to use, user satisfaction, and net benefits.

H14–There is a positive relationship between system quality, knowledge quality, service quality, intent to use, and net benefits of LCWU aKMS.

H15–There is a positive relationship between system quality, knowledge quality, service quality, user satisfaction, and net benefits of LCWU aKMS.

The proposed CKMF depicted in Fig. 4, consists of five layers that cover the overall flow and management of knowledge in LCWU aKMS software. It is customized for agriculture domain, the purpose of customizing the CKMF is to overcome the challenges in agriculture domain presented in the previous section by replacing the extension agent with a CKM system. The Internet, computers, and smartphones are now widely available. The proposed system aims to automate assisting the agricultural sector.

Using the proposed solution, a farmer can directly communicate with an expert. Moreover, a farmer can access stored information to get help and share experiences with other farmers using this platform. At the other end, experts can view and answer farmers’ queries, and also view and update the help materials provided to system users. LCWU aKMS was developed based on CKMF. Screenshots of LCWU aKMS are provided in Appendix B of this paper. The prime roles in this system are Farmer (Person who is the requester of information) and Expert (Individual who works as facilitator for the farmer).

Layers of CKMF and their customization for rice leave disease detection are discussed below.

The interface layer includes functions related to connection, access, and transformation controls. This layer forms the gateway, communication with the external world and the system is conducted through this layer. It contains settings for the user and device controls. This layer performs the functions of User Authentication, Portal Interface, Role-Based Access Control (RBAC), and Transformation Characteristics.

Customized CKMF Interface Layer: Using the user authentication feature, a farmer can be registered and subsequently use the login details to access the system. The portal interface is used by the farmer to interact with the system, where farmers can view and utilize the LCWU aKMS functionality with the interfaces. The RBAC functionality validates farmer and expert credentials by verifying user names and passwords. The transformation characteristics help to vary the ways it presents information on various devices, including smartphones, laptops, and desktop systems.

This layer provides the primary services of the framework. These services facilitate communications, KM, searches, and data analyses. The services layer holds the KM functionality, with all tasks (i.e., acquisition, generation, elaboration, storage, and sharing) related to knowledge are managed by this layer. The services layer provides primary services of Collaboration Services, Knowledge Management Services (knowledge acquisition, knowledge generation, knowledge elaboration, knowledge storage, and knowledge sharing), Data Search Services, and Data Analysis Services.

Customized CKMF Services Layer: Collaboration services involve communicating and sharing ideas for better performance and education of stakeholders. These tasks are achieved by implementing the following features in LCWU aKMS software (based on CKMF): Chatting (two-way communications), Sharing Experiences (Blogs), News Alerts (Broadcasting), and downloadable help documents provided for user guidance. Knowledge Management Services form the primary component of a KM system. The image acquisition step includes the process of gathering infected rice leaf images to implement the proposed workflow. Farmers can post problem descriptions along with images of infected rice leaves. Pre-processing is performed to remove noise from the rice leaf images. Diseases are identified based on feature matching by constructing a histogram of a new image and comparing it to histograms of already learned images. The system decides to suggest a cure based on the identified disease information shared by experts. Precautionary measures and cures are gathered from agricultural experts. The resulting knowledge is stored in the knowledge base for future reference and sharing. Disease data in the knowledge base are subject to ongoing updates based on experts’ recommendations. This feature facilitates sharing of previously-stored knowledge, supporting continuous improvements, and helping users make decisions. LCWU aKMS has the ability to new learn images, store their information and answer the farmer queries using this information. Moreover, it can learn new diseases after expert verification. This process is explained in Algorithm 1.

The farmer may request identification of a disease in a rice leaf, by providing an image of the leaf as input to the system. After receiving a request from the framer, the system checks its knowledge base for a matching answer. The information retrieved from the knowledge base is then transferred to an expert for approval. The information received from the expert is refined and stored in the knowledge base for future referencing and is also sent to the farmer as the answer to the query. Algorithm 2 shows this process.

Other requests from farmers could be to obtain help material or share ideas with co-farmers. Farmers can also generate feedback after getting a response from the system, which will help improve the system and data analysis.

Search Data Services help users locate the information of their interest, LCWU aKMS facilitates searching for matches for a given diseased leaf and searching for the cure for the identified disease

Analyse Data Services help users to analyse the collected knowledge, with the system generating reports based on various filters. LCWU aKMS software provides analysis reports based on the following criteria:

• By disease, producing a report for occurrences of a certain disease in all regions of Punjab

This layer contains Metadata, Taxonomy, Knowledge Structure, and Content Management. It holds all content-related tasks performed by the CKMF. The objective is to ensure the effective arrangement of information in the knowledge base to make it a useful source of knowledge for users.

Customized CKMF Content Layer: Metadata refers to explanations of valid data formats, showing the types of data that will be created and may include keywords, like crop name, the disease, and the region. Taxonomy involves business rules for storing and managing data; for LCWU aKMS, data are classified into images, diseases, and documents. Knowledge Structure indicates how knowledge is maintained and in the developed software, knowledge is arranged by disease. Content Management is the system’s general data management, including rules for validating new data, updating existing records, handling unwanted entries, obtaining and managing supporting materials. This layer is responsible for the arrangement of images and analysis results. It manages the storage of new images and controls the movement of learned images to trained data storage from new images storage location.

Physical layer contains settings for physical resources and search, filtering, and resource management services. Internet is used as the backbone for all information flows. This layer controls Repository Filters, Security Services, Information Services, and Asset Management-related features.

Customized CKMF Physical Layer: Repository Filters perform checks on data to be stored and retrieved from the system database. It controls the image matching process by responding to queries for images of interest. Security Services consists of specific checks, such as the username and password being authenticated for system usage. Information Services control information flows in the system, consisting of extraction of valid data, transforming extracted data into the desired form, and communicating answers to experts and users. Asset Management is the physical management of system data. LCWU aKMS manages images in the training and test datasets, feedback, answered and unanswered queries, help documents and expert opinions as assets.

The storage layer is responsible for the management of storage locations that hold the framework’s critical resources. It manages User Space (individuals, groups, and organizations) and Physical Repositories.

Customized CKMF Storage Layer: User space is managed for the following three user types:

Individual-Storage space allocated to a single user, such as a farmer or an expert.

Groups-Storage space is allotted for a group of experts and farmers, with members permitted to share stored resources.

Organization-Could be allocated to farmers and experts working in the same organization.

Physical Repositories hold the actual resources stored in databases. LCWU aKMS holds the following resources in repositories: help documents, messages, leaf images, feedback, discussion forums, user profiles, and user storage.

This section provides an overall success measure for the software and hence evaluates the CKMF from the KM perspective. The responses collected are statistically analysed to evaluate and assess LCWU aKMS. Tab. 2 lists the demographic information collected from the respondents.

Respondents were asked questions to assess the existing procedure (before using LCWU aKMS) of extension agent support. Tab. 3 provides the collected information, including frequency of extension agent visits, satisfaction with extension agent, and availability of online help.

The results showed that most extension agents visit three times a month and farmers are not satisfied with the services they provide. Moreover, the results also indicate online help is not available for farmers.

The reliability of the instrument is evaluated by calculating the internal consistency of the instrument using Cronbach’s Alpha. It is used to show how closely related a set of items are as a group. According to [48], 0.7 is the recommended value for Cronbach’s Alpha, but this value could be 0.6 for exploratory research [49]. Tab. 4 gives Cronbach’s Alpha for constructs used for CKMF evaluation, its value varies from 0.601 to 0.845, satisfying the internal consistency of the instrument items used to evaluate the CKMF.

Pearson’s correlation was used to evaluate the linear relationships, direction, and strength of the associations among variables. Tab. 5 shows Pearson’s correlation, the results show that a strong direct association exists among the constructs in this study. The 0.01 level shows the correlation coefficients are statistically significant.

The research work presented in this paper is evaluated against multiple aspects. The discussion section provides arguments about the designed framework i.e., CKMF for the following three aspects:

• Accuracy comparison of rice leaf disease identification module of LCWU aKMS with existing techniques

This section discusses KM features found in the gap analysis for CKMF. KM could occur at different levels; the communication and sharing of ideas make it collaborative in nature. The CKMF supports collaboration at all three levels (individual, team, and organizational). The literature examined showed that there are four modes of knowledge transfer: socialization (tacit to tacit), externalization (tacit to explicit), internalization (explicit to tacit), and combination (explicit to explicit). CKMF aims to address all four types. Tab. 6 indicates the functions of the LCWU aKMS, showing all four knowledge transfer modes.

KM is aided by the IT infrastructure of the organization for effective processing. The technology infrastructure is a combination of data processing, storage, communication technologies, databases, servers, and so on [14]. The final layer of the CKMF framework, referred to as the ‘Storage Layer’, holds the physical repositories along with other components. It contains the actual resources physically in the databases, including user profile information, help material, user requests, system/expert answers, chats, blogs, and data.

Information communication technology (ICT) tools refer to digital infrastructures that include devices, the internet, and other computing facilities. CKMF framework supports ICT tools, while the developed test case system (LCWU aKMS) supports both mobile and desktop devices. ICT tools are also involved in the image processing phase. Farmers can use their mobiles to take images of the infected leaves and send to system via internet. The help material provided for farmers is downloadable and printer-friendly. Several software features are mapped to CKMF framework including security, built-in help, and clearly defied context. Furthermore, the designed system is flexible, as new features can be easily added. The collaboration features help people communicate and share their ideas, experiences, and problems. This research’s goal was to develop a KM framework for collaborative environments. LCWU aKMS supports wikis, blogs, and social software features. Users can share information and data and communicate with each other. Tab. 7 presents a summarized view highlighting various features of the proposed CKMF against the parameters used to summarize existing KM frameworks stated in the related work section above.

6.2 Evaluation and Assessment of LCWU AKMS Software Using Evaluation Model and Hypotheses

The research questions are analysed by articulating fifteen hypotheses. The results indicate that all hypotheses are well supported. The empirical results of this study indicate a significant relationship among the six constructs extracted from the Jennex Olfman KM success model. The results were consistent with previous IS success model research [50–52], and the literature provides quite strong support for the designed hypotheses [44,46,53,54]. Regression analysis was performed to examine the research hypotheses. Regression analysis is used to measure the relationship between independent variables and dependent variables [55]. Tab. 8 portrays hypotheses are supported by analysing the coefficient of determination (R2). Independent and dependent variables are also stated for each hypothesis to understand the relationships.

The survey questionnaire was developed based on hypotheses (adapted from Jennex Olfman model) generated to check the effectiveness of the designed system. The evaluation model depicted in Fig. 3 supports the proposed hypothesis. Moreover, Tab. 8 shows that the hypotheses are supported by the results of the survey estimating the usefulness of LCWU aKMS.

Leaf disease detection is performed by many researchers, this section lists a few of such researches and compares the results with the accuracy of LCWU aKMS rice leaf disease detection technique. LCWU aKMS has shown 94.16% accuracy for rice leaf disease detection. Tab. 9 compares the existing and proposed methodologies.

KM has proven its beneficial worth by providing knowledge assets for individuals and organizations. KM is an important multidisciplinary area that contributes towards the integration of knowledge in existing practices by engaging individuals, communities, organizations both related to the private and government sector. Collaborative features coupled with KM features form a strong combination forming a KM system having collaboration features of communication and information sharing. Frameworks play a dynamic role in developing any software system. Researchers have designed several KM frameworks focusing on aspects like critical factors, knowledge level, software features, etc. This study focuses on presenting a generic collaborative KM framework termed CKMF, based on existing frameworks’ gap analysis. The Paper also provides evaluation for CKMF using three different aspects. The proposed CKMF framework is customized for implementation in the agriculture domain by taking a case study of Pakistan. The KM is well received in various areas such as education, medical, software development, etc. However, certain disciplines still need to integrate KM practices and frameworks to refine their work practices and acquire results that contribute towards economic growth. In this regard, agriculture is an important area particularly in the context of developing countries where a large population is associated with agriculture.

The empirical evidence in the current study suggests that an increase in system quality, knowledge quality, and service quality increases intent to use and user satisfaction, which increases the net benefit of the KMS. The research work provided practical implementation of the stated concepts of CKMF. LCWU aKMS accurately detected the diseased spots present (if any), classified the type of the disease affecting the leaf, and suggested cures were provided for the diseases identified. The disease detection module is evaluated using the precision and recall method and found to be 94.16% accurate. The results obtained may help farmers be more effective in decision making, more efficiently protecting their rice crops from substantial damage. The system could replace the work of extension agents, making it a cost and time-effective initiative for farmer betterment.

The methodology presented can aid in precision agriculture by forming the basis for rice disease KM. LCWU aKMS provide a platform for farmers for collaborative communication and sharing. CKMF could be customized for KM activities in other domains like education, medicine, media, software houses, and others.

Conflicts of Interest: The authors declare that they have no conflicts of interest to report regarding the present study.

References

1. M. Kavalić, M. Nikolić, D. Radosav, S. Stanisavljev and M. Pečujlija, “Influencing factors on knowledge management for organizational sustainability,” Sustainability (Switzerland), vol. 13, no. 3, pp. 1–18, 2021. [Google Scholar]

3. A. Kothari, D. Rudman, M. Dobbins, M. Rouse, S. Sibbald et al., “The use of tacit and explicit knowledge in public health: A qualitative study,” Implementation Science, vol. 7, no. 1, pp. 20, 2012. [Google Scholar]

4. F. O. Omotayo, “Knowledge management as an important tool in organisational management: A review of literature,” Library Philosophy and Practice, vol. 1, pp. 1238, 2015. [Google Scholar]

5. E. A. Smith, “The role of tacit and explicit knowledge in the workplace,” Journal of Knowledge Management, vol. 5, no. 4, pp. 311–321, 2001. [Google Scholar]

6. K. Bashir and M. Rehman, “Comparative analysis of knowledge management frameworks,” International Journal of Computer Science and Information Security, vol. 14, no. 12, pp. 80–87, 2016. [Google Scholar]

7. H. Ayatollahi and K. Zeraatkar, “Factors influencing the success of knowledge management process in health care organisations: A literature review,” Health Information & Libraries Journal, vol. 37, no. 2, pp. 98–117, 2019. [Google Scholar]

8. S. Venkatraman and R. Venkatraman, “Communities of practice approach for knowledge management systems,” Systems, vol. 6, no. 4, pp. 36, 2018. [Google Scholar]

10. J. Mostert and M. Snyman, “Knowledge management framework for the development of an effective knowledge management strategy,” South African Journal of Information Management, vol. 9, no. 2, pp. 894–902, 2007. [Google Scholar]

11. B. Dave and L. Koskela, “Collaborative knowledge management-A construction case study,” Automation in Construction, vol. 18, no. 7, pp. 894–902, 2009. [Google Scholar]

12. F. Corno, P. Reinmoeller and I. Nonaka, “Knowledge creation within industrial systems,” Journal of Management and Governance, vol. 3, no. 4, pp. 379–394, 1999. [Google Scholar]

13. K. M. Wiig, “Knowledge management: An emerging discipline rooted in a long history,” Knowledge Horizons: The Present and the Promise of Knowledge Management, vol. 3, pp. 3–26, 2000. [Google Scholar]

15. B. V. den Hooff and J. A. de Ridder, “Knowledge sharing in context: The influence of organizational commitment, communication climate and CMC use on knowledge sharing,” Journal of Knowledge Management, vol. 8, no. 6, pp. 117–130, 2004. [Google Scholar]

16. S. Wang and R. A. Noe, “Knowledge sharing: A review and directions for future research,” Human Resource Management Review, vol. 20, no. 2, pp. 115–131, 2010. [Google Scholar]

17. I. Nonaka, “Dynamic theory knowledge of organizational creation,” Organization Science, vol. 5, no. 1, pp. 14–37, 1994. [Google Scholar]

18. M. Alavi and D. E. Leidner, “Review: Knowledge management and knowledge management systems: Conceptual foundations and research issues,” MIS Quarterly, vol. 25, no. 1, pp. 107–136, 2001. [Google Scholar]

19. T. Jacks, S. Wallace and H. Nemati, “Impact of culture on knowledge management: A meta-analysis and framework,” Journal of Global Information Technology Management, vol. 15, no. 4, pp. 8–42, 2012. [Google Scholar]

20. Z. O. Atkočiūnienė, “Knowledge management : Collaborative knowledge aspect,” Electronic Learning, Information and Communication: Theory and Practice, vol. 1, no. 1, pp. 53–63, 2014. [Google Scholar]

21. R. Michaelides, S. C. Morton, Z. Michaelides, A. C. Lyons and W. Liu, “Collaboration networks and collaboration tools: A match for SMEs?” International Journal of Production Research, vol. 51, no. 7, pp. 1–15, 2012. [Google Scholar]

22. T. Marlowe, V. Kirova, N. Jastroch and M. Mohtashami, “A classification of collaborative knowledge,” Systemics, Cybernetics and Informatics, vol. 9, no. 7, pp. 24–29, 2011. [Google Scholar]

23. N. Asmani and A. Ekadinata, “Toward green economic growth in south sumatra: Collaborative platform through a knowledge management approach,” in Proc. of IOP Conf. Ser. Earth Environ. Sci., South Sumatera, Indonesia, vol. 298, no. 1, 2019. [Google Scholar]

24. A. Ahmed and A. Ahsan, “An integrated organizational knowledge management framework (IOKMF) for knowledge creation and usage,” Journal of Strategy & Performance Management, vol. 2, no.1, pp. 17–30, 2014. [Google Scholar]

27. H. Chitto, B. M. Nowbutsing and R. Ramchurn, “Knowledge management : Promises and premises,” Global Journal of Management and Bussiness Research, vol. 10, no. 1, pp. 123–131, 2010. [Google Scholar]

28. D. Karimzadegan, J. Tanha and E. Majd, “Providing a comprehensive knowledge management model,” Journal of Research in International Business and Management, vol. 1, no. 6, pp. 155–163, 2011. [Google Scholar]

29. B. Shakya, F. Schneider, Y. Yang and E. Sharma, “A multiscale transdisciplinary framework for advancing the sustainability agenda of mountain agricultural systems,” Mountain Research and Development, vol. 39, no. 3, pp. A1–A7, 2020. [Google Scholar]

30. S. Liu, L. Olfman and T. Ryan, “Knowledge management system success: Empirical assessment of a theoretical model,” International Journal of Knowledge Management, vol. 1, no. 2, pp. 68–87, 2005. [Google Scholar]

31. M. Abbas, T. E. Lodhi, K. M. Aujla and S. Saadullah, “Agricultural extension programs in Punjab, Pakistan,” Pakistan Journal of Life and Social Sciences, vol. 7, no. 1, pp. 1–10, 2009. [Google Scholar]

34. S. Fountas, G. Carli, C. G. Sørensen, Z. Tsiropoulos, C. Cavalaris et al., “Farm management information systems: Current situation and future perspectives,” Computers and Electronics in Agriculture, vol. 115, pp. 40–50, 2015. [Google Scholar]

35. L. Ali and A. Avdic, “A knowledge management framework for sustainable rural development: The case of gilgit-baltistan, Pakistan,” Electronic Journal of Knowledge Management, vol. 13, no. 2, pp. 103–116, 2015. [Google Scholar]

37. F. Nnadi, J. Chikaire, C. Atoma, H. A. Egwuonwu and J. E. Echetama, “ICT for agriculture knowledge management in Nigeria: Lessons and strategies for improvement,” Science Journal of Agricultural Research & Management, vol. 2012, pp. 8, 2012. [Google Scholar]

41. M. A. Baloch and G. B. Thapa, “Review of the agricultural extension modes and services with the focus to balochistan, Pakistan,” Journal of the Saudi Society of Agricultural Sciences, vol. 18, no. 2, pp. 188–194, 2019. [Google Scholar]

43. M. Jennex and L. Olfman, “A knowledge management success model : An extension of DeLone and McLean’s IS success model,” in Proc. of AMCIS, Tampa, FL, USA, 2003. [Google Scholar]

44. M. H. Wang and T. Y. Yang, “Investigating the success of knowledge management: An empirical study of small-and medium-sized enterprises,” Asia Pacific Management Review, vol. 21, no. 2, pp. 79–91, 2016. [Google Scholar]

45. H. M. Al-Hattami, “Validation of the D&M IS success model in the context of accounting information system of the banking sector in the least developed countries,” Journal of Management Control, vol. 32, no. 1, pp. 127–153, 2021. [Google Scholar]

46. M. E. Jennex and L. Olfman, “A model of knowledge management success,” International Journal of Knowledge Management, vol. 2, no. 3, pp. 51–68, 2006. [Google Scholar]

47. M. E. Jennex, “Re-examining the Jennex Olfman knowledge management success model,” in Proc. of Hawaii Int. Conf. Syst. Sci., Hawaii, USA, pp. 4375–4384, 2017. [Google Scholar]

48. F. A. Mir and A. H. Pinnington, “Exploring the value of project management: Linking project management performance and project success,” International Journal of Project Management, vol. 32, no. 2, pp. 202–217, 2014. [Google Scholar]

49. L. L. Chan and N. Idris, “Validity and reliability of the instrument using exploratory factor analysis and cronbach’s alpha,” International Journal of Academic Research in Business and Social Sciences, vol. 7, no. 10, pp. 2222–6990, 2017. [Google Scholar]

50. P. B. Seddon, “A respecification and extension of the DeLone and McLean model of IS success,” Journal of Information Systems Research, vol. 8, no. 1, pp. 240–253, 1997. [Google Scholar]

51. A. Molla and P. Licker, “E-Commerce systems success: An attempt to partially extend and respecify the delone and mclean model of IS success,” Journal of Electronic Commerce Research, vol. 2, no. 4, pp. 131–141, 2010. [Google Scholar]

52. W. H. Delone and E. R. Mclean, “The DeLone and McLean model of information systems success: A ten-year update,” Journal of Management Information Systems, vol. 19, no. 4, pp. 9–30, 2003. [Google Scholar]

53. L. A. Halawi, R. V. McCarthy and J. E. Aronson, “An empirical investigation of knowledge management systems’ success,” Journal of Computer Information Systems, vol. 48, no. 2, pp. 121–135, 2007. [Google Scholar]

54. B. Armstrong and G. Fogarty, “Validation of a computer user satisfaction questionnaire to measure IS success in small business,” Journal of Research and Practice in Information Technology, vol. 37, no. 1, pp. 27–42, 2005. [Google Scholar]

56. S. Phadikar, J. Sil and A. Das, “Classification of rice leaf diseases based on morphological changes,” International Journal of Information and Electronics Engineering, vol. 2, no. 3, pp. 460–463, 2012. [Google Scholar]

57. C. K. Charliepaul, “Classification of rice plant leaf,” International Journal on Engineering Technology and Sciences–IJETS, vol. 1, no. 7, pp. 290–295, 2014. [Google Scholar]

58. N. R. Kakade and D. D. Ahire, “Real time grape leaf disease detection,” International Journal of Advance Research and Innovative Ideas in Education, vol. 1, no. 4, pp. 598–610, 2015. [Google Scholar]

59. S. S. Sannakki and V. S. Rajpurohit, “An approach for detection and classification of leaf spot diseases affecting pomegranate crop,” International Journal of Advance Foundation and Research in Computer, vol. 2, no. 1, pp. 317–327, 2015. [Google Scholar]

60. K. J. Mohan, “Detection and recognition of diseases from paddy plant leaf images,” International Journal of Computer Applications, vol. 144, no. 12, pp. 34–41, 2016. [Google Scholar]

61. A. N. H. Zaied, “An Integrated Success Model for Evaluating Information System in Public Sectors,” Journal of Emerging Trends in Computing and Information Sciences, vol. 3, no. 6, pp. 814–825, 2012. [Google Scholar]

Computer Systems Science & Engineering DOI:10.32604/csse.2022.022247
Article