submitted time 2024-04-03 Cooperative journals: 《农业图书情报学报》
Abstract： [Purpose/Significance] In order to explore the technological gaps in Chinese im-portant agricultural fields and predict the future trends of these gaps, this study investigates technology opportunity discovery in the embryonic and developmental stages from the per-spectives of technology gap discovery and technology fusion opportunity discovery, provid-ing consultation and suggestions for decision-makers on the technology development op-portunities for technology innovation. [Method/Process] First, TextRank method was used to mine information in abstracts of papers and patents in this paper, which is a key sentence embedding method. The sentence vector clustering method was applied to extract topic sen-tences of papers and patents. Second, comparative analysis of topic clustering was used to detect technology gaps. Third, semantic similarity networks and classification similarity networks were used to discover the theme directions, which are likely to develop into cross-domain research areas with these technology gaps. [Results/Conclusions] The experi-mental results indicate that the proposed method can identify technological gaps. Combined with expert analysis, the experimental results can show the current development status and predict the trends of genomics technology in the field of animal husbandry and veterinary medicine. At the same time, this study can provide methodological and data support for genomics technology think tanks in the field of animal husbandry and veterinary medicine in China. Specifically, China has published a large number of papers and patents, but the tech-nical architecture layout is not as complete as foreign countries. The topics of Chinese papers are more complete than those of Chinese patents. In addition, China lacks sufficient basic research support in the integration and association of multi-omics, and the technical conditions are also incomplete. The field of genetically modified (GM) breeding technology is also recognized as a technological gap in China. In addition, it is possible that GM breeding and whole genome association analysis, multi-omics integration and viral genome analysis of livestock and poultry will become new technological fusion points in the future. There are still drawbacks in this study: It still takes time and manpower to manually analyze and interpret the relationship between scientific papers and technological patents. In the future research, more automated methods will be designed to construct correlation comparison methods between two data objects. Additionally, there is still room for improvement in expert interpretation of clustering themes. In the future, more data can be considered to add label information, reducing manual annotation work while providing the possibility of increasing quantitative accuracy in the result validation section.
submitted time 2024-04-03 Cooperative journals: 《农业图书情报学报》
Abstract： [Purpose/Significance] In the era of big data, the volume of data is growing at an exponential rate. One of the most prominent areas affected by this growth is the field of agriculture. The use of agricultural knowledge graphs, which serve as key infrastructures for managing agricultural knowledge, has expanded significantly. However, as the number of nodes and relationships within these graphs increase, so too does their complexity. This complexity gives rise to new challenges in training and representing such large-scale knowledge graphs. It is therefore of great significance to investigate methods for speeding up the embedding process of agricultural knowledge graphs, while preserving their structural integrity and minimizing resource consumption. This research embarks on a novel exploration to address this issue. It stands out from previous studies by concentrating on a hierarchical representation model for agricultural knowledge graphs. The potential impacts of this research on propelling the advancement of the field and on addressing significant real-world problems are substantial. [Method/Process] To confront this challenge, we propose a hierarchical representation model for agricultural knowledge graphs rooted in the HARP framework. Our model leverages the inherent hierarchical features of the agricultural knowledge graph. It incorporates an improved random walk strategy based on relational paths to semantically model relationship objects within the agricultural knowledge graph. This innovative approach effectively retains the hierarchy and asymmetrical relationship structure of the nodes in the graph, setting our work apart from previous research. The validity of our proposed model is fortified by a strong foundation of theoretical and empirical evidence. [Results/Conclusions] Our experimental results reveal several key findings. First, the hierarchical random walk with path (HRWP) model using the LEIDEN algorithm can preserve the spatial structure more effectively and converge more quickly to the maximum modularity, in comparison to the HARP framework. Second, the fusion model employing HRWP takes less training time than the total training time of both models combined, without significantly affecting the time complexity of the original algorithm. Third, we observed that when traditional algorithms are integrated with HRWP, there is an average improvement of 2% across various indicators, with a substantial enhancement in non-neural network models. Therefore, our proposed model not only accurately represents the agricultural knowledge graph but also effectively reduces the training time. Despite the promising outcomes of our study, there remain areas of potential improvement. One such area is the need for a more detailed discussion on the hierarchical nature of relationship objects in future research. This provides potential avenues for future exploration in this field. The findings of this research carry profound implications for the development of agricultural knowledge management systems, offering an effective approach to handle the burgeoning complexity of knowledge graphs.
submitted time 2024-04-03 Cooperative journals: 《农业图书情报学报》
Abstract： [Purpose/Significance] This study aims to complement the shortcomings existing in the current research, such as the limited research on the micro-level of enabling rural revitalization and the unclear impact of knowledge services on rural revitalization, in order to strengthen the understanding of the mechanism and path of enabling rural revitalization by agricultural intelligent knowledge services, and to provide theoretical reference and practical solutions for better enabling rural revitalization by the knowledge services and better developing agricultural intelligent knowledge services. [Method/Process] This paper first reviews the literature to understand the current research status of enabling rural revitalization and knowledge services affecting rural revitalization, and then determines the research focus. Based on the three perspectives of data, application and service, this paper explores the enabling basis of agricultural intelligent knowledge services, analyzes the enabling mechanism from the three aspects of agricultural production, farmer life and rural governance, clarifies the necessity and possibility of agricultural intelligent knowledge services enabling rural revitalization, and constructs the internal mechanism of the knowledge services enabling rural revitalization. Moreover, the predicament of agricultural intelligent knowledge services enabling rural revitalization is analyzed from the internal and external aspects, and the performance and impact of these dilemmas are analyzed. Finally, according to the analysis results, the optimization path is proposed from the three dimensions of enabling demand, scene and effect to alleviate the problem. [Results/Conclusions] Agricultural knowledge services have entered the intelligent stage, presenting the characteristics of intelligence, personalization and precision, and have established a good foundation for enabling rural revitalization in terms of data, application and service. By enabling the development of new production modes, the transformation of farmers' multiple identities, and the equal participation of multiple subjects in governance, digital intelligence momentum will be injected into the comprehensive promotion of rural revitalization. However, from the internal perspective of agricultural intelligent knowledge services, there are some shortcomings such as obvious user group bias and the lack of service content. From the external point of view of the knowledge services, there are the challenges of unbalanced infrastructure and low user literacy. These deficiencies and challenges limit the enabling role of agricultural intelligent knowledge services in rural revitalization, which will limit the maximum play of the enabling effects. Accordingly, this paper provides the corresponding solution. First, in-depth research could be carried out on rural revitalization stakeholders and strategic content to clarify the empowerment needs. Second, integrated service platform could be built to integrate and empower scenarios. Third, the promotion model could be constructed to enhance the enabling effect to alleviate the empowerment dilemma.
submitted time 2024-04-03 Cooperative journals: 《农业图书情报学报》
Abstract： [Purpose/Significance] The ancient Chinese agricultural books are the main carrier of traditional agricultural experience, and represent the productivity and the essence of agricultural history in China. The value of agricultural knowledge in them has not disappeared with the progress of the times, and still has practical guidance for the problems that arise in modern agriculture. However, the ancient Chinese agricultural books are written in ancient Chinese, which are obscure and without punctuation, making them difficult to use. Semantic retrieval is a retrieval method that automatically queries and extracts relevant information from information sources at the semantic level. It can accurately capture the true intention behind user problems and conduct searches based on it, and thereby it is capable of returning more accurate and the most consistent results to users. However, currently most relevant research only focuses on major languages, and there is insufficient research on sentence embedding in ancient Chinese prose. In order to fill the gap in the field and provide scholars with more convenient methods for retrieving ancient agricultural knowledge and tracing ancient agricultural knowledge, this study is based on comparative learning methods to construct a semantic retrieval model that can automatically return the most relevant ancient agricultural paragraph with input, using vernacular Chinese as the query. [Method/Process] SikuBERT, which is based on Siku Quanshu as the training corpus, is used as the basic model. Based on the method of comparative learning, the model is continued to be trained using the self-built ancient agricultural dataset, and a semantic retrieval model that can support the use of vernacular as a query and return the ancient agricultural paragraphs most similar to the query semantics is obtained. [Results/Conclusions] The Spearman coefficient of the ancient agricultural text semantic retrieval model can achieve 86.51% performance on the test set, which is a certain degree of improvement compared to the baseline model's 83.69% performance on the test set. The recall situation on the self built ancient agricultural literature retrieval test set has been improved to a certain extent compared to the baseline model, and the model can have good retrieval results on ancient agricultural literature. However, semantic retrieval models usually require relevant semantic similarity datasets or semantic matching datasets for training. Due to the lack of large-scale and pure ancient Chinese data in the field of ancient agricultural literature, and the high cost of constructing relevant datasets requiring personnel with high-standard relevant professional qualifications, this experiment used a self-built dataset for training, which is limited by the quantity and quality of ancient agricultural language corpus data. The current semantic retrieval model for ancient agricultural literature is still not as effective as expected. In the future, we will search for training methods suitable for small samples, such as transfer learning based on cross language pre-training models to improve the retrieval performance.
Subjects: Library Science，Information Science >> Library Science submitted time 2023-09-05 Cooperative journals: 《农业图书情报学报》
Abstract： Purpose/Significance Knowledge Graph has become a major research hotspot in the era of artificial intelligence due to its ability to provide a new means of organization and representation of knowledge. As the field continues to evolve, numerous scholars have proposed advanced algorithms and technologies for each core stage of constructing a knowledge graph, and many large domestic and foreign enterprises have also developed their independent knowledge graph management systems. However, the majority of these graph tools developed are designed for commercial use and are often too expensive and difficult to deploy locally for small and medium-sized research teams. This presents a challenge for information organizations such as research libraries with massive resources, which require a more adaptable, universal, and efficient tool to build and manage knowledge graphs. To meet this need, it is important to develop an open-source, user-friendly, and customizable knowledge graph management system that can be easily deployed by small and medium-sized research teams. Method/Process In summary, this article offers a thorough and informative analysis of six mainstream knowledge graph management systems, both domestically and internationally. It delves into the unique characteristics of each system within the business process and provides an in-depth comparative analysis based on several important factors, including system functionality, technology selection, open-source availability, and application domains. The article refers to the standard construction process of knowledge graphs and highlights the platform characteristics of each system during the construction process while also examining their limitations based on current data characteristics. In response to practical needs, the article focuses on multi-path, multi-engine, distributed, and collaborative construction, integrating advanced graph algorithms and considering a well-developed underlying graph storage strategy. Results/Conclusions As a result袁the article presents an in-depth analysis of the construction model for a collaborative development and management system of an integrated knowledge graph. It not only investigates the current state of knowledge graph management systems but also proposes novel optimization ideas. These ideas include distributed collaborative construction, which allows for simultaneous contributions from multiple sources, and parallel management of multiple graphs, enabling efficient organization and retrieval. Additionally, some suggestions are put forward: developing multi-path knowledge extraction techniques to enhance the knowledge acquisition process, and using specialized multi-graph storage engines for optimized storage and retrieval. Last, the article emphasizes the importance of incorporating cross-media and multimodal knowledge into the graph for a comprehensive representation of information.
Subjects: Agriculture, Forestry,Livestock & Aquatic Products Science >> Basic Disciplines of Agriculture submitted time 2023-08-14 Cooperative journals: 《智慧农业（中英文）》
Abstract： [Objective] Desert vegetation is an indispensable part of desert ecosystems, and its conservation and restoration are crucial. Accurate identification of desert plants is an indispensable task, and is the basis of desert ecological research and conservation. The complex growth environment caused by light, soil, shadow and other vegetation increases the recognition difficulty, and the generalization abili‐ty is poor and the recognition accuracy is not guaranteed. The rapid development of modern technology provides new opportunities for plant identification and classification. By using intelligent identification algorithms, field investigators can be effectively assisted in desert plant identification and classification, thus improve efficiency and accuracy, while reduce the associated human and material costs. [Methods] In this research, the following works were carried out for the recognition of desert plant: Firstly, a training dataset of deep learning model of desert plant images in the arid and semi-arid region of Xinjiang was constructed to provide data resources and basic support for the classification and recognition of desert plant images.The desert plant image data was collected in Changji and Tacheng region from the end of September 2021 and July to August 2022, and named DPlants50. The dataset contains 50 plant species in 13 families and 43 genera with a total of 12,507 images, and the number of images for each plant ranges from 183 to 339. Secondly, a migration integration learning-based algorithm for desert plant image recognition was proposed, which could effectively improve the recognition accuracy. Taking the EfficientNet B0−B4 network as the base network, the ImageNet dataset was pre-trained by migration learning, and then an integrated learning strategy was adopted combining Bagging and Stacking, which was divided into two layers. The first layer introduced K-fold cross-validation to divide the dataset and trained K sub-models by borrowing the Stacking method. Considering that the output features of each model were the same in this study, the second layer used Bagging to integrate the output features of the first layer model by voting method, and the difference was that the same sub-models and K sub-models were compared to select the better model, so as to build the integrated model, reduce the model bias and variance, and improve the recognition performance of the model. For 50 types of desert plants, 20% of the data was divided as the test set, and the remaining 5 fold cross validation was used to divide the dataset, then can use DPi(i=1,2,…,5) represents each training or validation set. Based on the pre trained EfficientNet B0−B4 network, training and validation were conducted on 5 data subsets. Finally, the model was integrated using soft voting, hard voting, and weighted voting methods, and tested on the test set. [Results and Discussions] The results showed that the Top-1 accuracy of the single sub-model based on EfficientNet B0 network was 92.26%~93.35%, the accuracy of the Ensemble-Soft model with soft voting, the Ensemble-Hard model with hard voting and the Ensemble-Weight model integrated by weighted voting method were 93.63%, 93.55% and 93.67%, F1 Score and accuracy were comparable, the accuracy and F1 Score of Ensemble-Weight model integrated by weighted voting method were not significantly improved compared with Ensemble-Soft model and Ensemble-hard model, but it showed that the effect of weighted voting method proposed in this study was better than both of them. The three integrated models demonstrate no noteworthy enhancements in accuracy and F1 Score when juxtaposed with the five sub-models. This observation results suggests that the homogeneity among the models constrains the effectiveness of the voting method strategy. Moreover, the recognition effects heavily hinges on the performance of the Efficient‐ Net B0-DP5 model. Therefore, the inclusion of networks with more pronounced differences was considered as sub-models. A single sub-model based on EfficientNet B0−B4 network had the highest Top-1 accuracy of 96.65% and F1 Score of 96.71%, while Ensemble-Soft model, Ensemble-Hard model and Ensemble-Weight model got the accuracy of 99.07%, 98.91% and 99.23%, which further improved the accuracy compared to the single sub-model, and the F1 Score was basically the same as the accuracy rate, and the model performance was significant. The model integrated by the weighted voting method also improved accuracy and F1 Score for both soft and hard voting, with significant model performance and better recognition, again indicating that the weighted voting method was more effective than the other two. Validated on the publicly available dataset Oxford Flowers102, the three integrated models improved the accuracy and F1 Score of the three sub-models compared to the five sub-models by a maximum of 4.56% and 5.05%, and a minimum of 1.94% and 2.29%, which proved that the migration and integration learning strategy proposed in this paper could effectively improve the model performances. [Conclusions] In this study, a method to recognize desert plant images in natural context by integrating migration learning and integration learning was proposed, which could improve the recognition accuracy of desert plants up to 99.23% and provide a solution to the problems of low accuracy, model robustness and weak generalization of plant images in real field environment. After transferring to the server through the cloud, it can realize the accurate recognition of desert plants and serve the scenes of field investigation, teaching science and scientific experiment.
Subjects: Other Disciplines >> Synthetic discipline submitted time 2023-03-31 Cooperative journals: 《农业图书情报学报》
Abstract： [Purpose/Significance] At present, most of the training models used in the research of news classification are non-active learning. There are common problems about these models, including data cannot be labeled immediately and the labeling cost is too high, which also hinders the analysis of agricultural news. Especially because of the explosive growth of news data in the network era, it is more difficult to label data, train supervised text classification models, and screen relevant news in the field of agriculture from diversified online news sources. In order to solve this problem, the most commonly used pool based active learning or deep active learning technique is used to select more valuable and representative data from unlabeled data for manual labeling, and construct labeled data sets to improve the efficiency and effect of news classification and agricultural news mining. [Method/Process] The commonly used machine learning models for text classification, such as random forest classifier, polynomial naive Bayes classifier and logistic regression classifier, were combined with the active learning method with the lowest confidence to analyze the effect, and the BERT model was combined with the three sampling strategies of discriminative active learning, deep Bayes active learning and lowest confidence for deep active learning training. On the news corpus of 19 847 samples crawled and cleaned by crawler technology from Sina and other news websites, aiming at screening agricultural related news from diversified news samples of various topics, the iterative experiment of adding 30 samples per round was tested to check the improvement effect of F1 score under various method combinations with the increase of the number of annotation. In addition, the representativeness and diversity of the samples selected by the sampling function of each method in the deep active learning method of the BERT model were compared, so as to understand the characteristics of each strategy and provide inspiration for the selection and improvement of Al strategy in the future. In addition, this paper also analyzed how much labeling cost can be saved by using the proposed method. [Results/Conclusions] When comparing a variety of machine learning models, it is found that although the gradient boosting tree and support vector machine classifier have high accuracy, they are not suitable for active learning because of their low efficiency in text data processing of large-scale high-dimensional data. After combining other machine learning models and the BERT model and training text models with the corresponding active learning or deep active learning methods, it is found that the application of active learning method can significantly improve the training process of each model. Among them, the BERT model, combined with discriminative active learning sampling function, has the best news text classification effect and the lowest annotation data requirements. The representativeness and diversity of the samples selected by discriminative active learning sampling function are also the highest, which explains the source of the advantages of this method. It can also be found that for the same task model, the higher the accuracy of classification is required, and the active learning method can save more annotation cost than non-active learning.
Subjects: Other Disciplines >> Synthetic discipline submitted time 2023-03-31 Cooperative journals: 《农业图书情报学报》
Abstract： [Purpose/Significance] This paper aims to study the structure-utility relationship of descriptive information of scientific data to provide a new perspective for the theoretical study of scientific data description and a reference for the best description of agricultural scientific data in the digital environment. [Method/Process] Based on information processing theory, the lens model, the probabilistic mental model theory and the adaptive decision-making behavior framework, the relationship model between descriptive information structure and informing utility was constructed. A situational experiment was designed according to the model. In this study, 47 postgraduates from 14 institutes were invited for quasi-experimental observation by using qualitative and quantitative methods such as eye-tracking, semi-structured interview and questionnaire. First, this study used a semi-structured interview to obtain a user's cognitive interpretation of fixation points and collected the descriptive items of agricultural scientific data and their use frequency by encoding the interview text. Second, this study combined descriptive item usage path coding and user judgment confidence to obtain the combination of descriptive items with high utility. Finally, the study used multiple regression analysis to identify the descriptive items with high utility and their predictive ability, and analyzed the impact of data literacy and data utilization type on the utility of descriptive items. [Results/Conclusions] The study identified 42 descriptive items of 11 categories of agricultural scientific data and their usage characteristics. Among them, the top 5 frequently used descriptive items were subject, data, overall description, source and data production information, which played an important role in user relevance judgment. Then this study identified the combination of descriptive items with high utility and found that users' use patterns of descriptive items were diverse. Compared with making a judgment with "relevant" result, users often needed less information to achieve a high level of confidence when making an "irrelevant" judgment. This study also found that the descriptive items with high utility include source, data, use and evaluation, and data production information. It is determined that user data literacy and data utilization purpose were the influencing factors of descriptive information utility, and the effects of the two factors were preliminarily analyzed. Based on this research, the paper put forward some suggestions for improving agricultural scientific data metadata and scientific data sharing. In the future, this study will be repeated in groups with different academic backgrounds and data literacy levels, so as to enhance the generalization ability of research conclusions and construct a more effective structure of scientific data descriptive information.
Subjects: Other Disciplines >> Synthetic discipline submitted time 2023-03-31 Cooperative journals: 《农业图书情报学报》
Abstract： [Purpose/Significance] The huge economic benefits of Bt insect resistant genes have attracted great attention of relevant research institutions and enterprises at home and abroad, and these genes and technologies are transformed into their own exclusive rights by using intellectual property rights. Through statistical analysis of Bt insect resistant gene patents collected in 126 countries and regions around the world, this paper provides a reference for China in the development and industrial application of Bt insect resistant genes. [Method/Process] Based on the Global Patent Database of Smart Bud (PatSnap), the title, abstract and claim of the patent were taken as the focus of the search, and the patent analysis method was used to statistically analyze the Bt insect-resistant transgenic patents collected in the past 30 years since 1992. In the process of Bt anti-insect gene research, a large number of patent data were produced. After data cleaning, the "Bt strain", "detection method", "Bt insecticidal reagent" and other patents unrelated to the anti-insect gene were eliminated, and 2,988 patents were obtained, all of which were invention patents. The research status, research hotspots and development trends of global Bt pest resistance genes were studied from the aspects of total patent applications and trend, gene family classification, national/regional distribution, key technology fields, technology flow, institutional competitiveness, patent legalization status, and compound trait patents. [Results/Conclusions] Results show that the global Bt insect-resistant gene technology patent research and development trends, research hotspots and distribution technology gradually remain stable, the trend of patent application and disclosure in China is growing at the same time as that in the world. China's patent applications account for an increasing proportion of the global patent applications. But compared with international patent quality, there is still a gap between China and other countries. Multinational companies are the main body of Bt pest resistance gene market competition, and Chinese applicants are mainly scientific research institutions and universities. The cry1 family is the main target of Bt pest resistance gene patent application. The crystallized protein of Bacillus is a hot topic at present. The United States is the exporting country of Bt insect-resistant gene technology, while China is still in the stage of technology input. The patent failure rate of Bt insect resistance gene is much lower than the average level in this field. There is great potential in the study of fusion genes and gene superposition composite traits. Finally, the development prospects of Bt insect resistance gene technology in China were discussed.
Subjects: Other Disciplines >> Synthetic discipline submitted time 2023-03-31 Cooperative journals: 《农业图书情报学报》
Abstract： [Purpose/Significance] Research on the demand and supply analysis of agricultural literature resources is of great significance to the collection construction of agricultural libraries, especially under the current new situation. The demand analysis of academic resources of agricultural science has always been a difficult task for agricultural special libraries because it covers a wide range of disciplines and involves the multiple cross and integrated applications of various disciplines with agriculture science. This paper innovates the methods of demand and supply analysis based on multi-level information organization and fine-grained data analysis, which would discover both the systematic demands and accurate demands of agricultural researchers, and help to adapt to the changes of agricultural science research paradigm. [Method/Process] This study took the Chinese Academy of Agricultural Sciences (CAAS) as an example, collecting its published foreign language journals and the corresponding cited references, as well as its usage data of foreign language electronic journals databases. Then this paper used the methods of citation analysis, subject organization, etc. to analyze the institutional demand characteristics that include literature language, document type, publication years, disciplines distribution, covered subjects and the utilized database of foreign-language periodical resources. The full-text supply rate of foreign-language electronic journals has been calculated to evaluate the satisfaction and the ability of supplying collection resources. The literature comprehensive utilization index has also been calculated to reflect the literature utilization capacity of CAAS's affiliated institutes by the methods of correlation analysis and multiple correlation coefficient weighting. [Results/Conclusions] The results show that the demands of foreign-language journals of CAAS involve not only classical agricultural disciplines but also some emerging and interdisciplinary fields, and there are big gaps in the level of literature supply and utilization among its affiliated institutes. The construction of literature resource supply system of CAAS should strengthen the demand analysis, enrich the subscription mode of full-text resources, and expand the non-full-text resources. As for the agricultural special libraries, the demand and supply analysis of agricultural science literature resources should follow the development trends of agricultural disciplines, taking into account the systematic and prospective demands, and carry out accurate, thematic and knowledge-based demand analysis in the relevant subject areas. Users can also be encouraged to participate in library collection construction, and diversified demand feedback channels should be provided for them. At the same time, resource construction quality and service efficiency evaluation should also be brought to the forefront.
Subjects: Other Disciplines >> Synthetic discipline submitted time 2023-03-31 Cooperative journals: 《农业图书情报学报》
Abstract： [Purpose/Significance] In the era of big data, people are flooded with massive information, and problems such as knowledge anxiety, adjustment of resource demand structure, and desire for high-quality information follow one after another. From the perspective of technical support, user interaction is rarely explored, but it is indispensable. The application in the field of infrastructure construction and business is relatively complete, but the user interaction of the knowledge service system is still insufficient: at the theoretical research level, there is a lack of summary of theoretical methods and systematic framework design; at the application practice level, there is a lack of systematic guidance. [Method/Process] The agricultural professional knowledge service system has relatively complete and representative user interaction, a large user base, and a high degree of retention, which is worthy of study, but it has certain shortcomings. The research on user interaction of agricultural knowledge system in this paper is mainly divided into the following three aspects. First, by sorting out the research status of user interaction at home and abroad, the user interaction framework of knowledge service system, namely human-computer interaction and interpersonal interaction, constitutes the basic research framework of this research. Second, based on this, using questionnaires and Baidu statistics this paper investigates the user demand of the agricultural professional knowledge service system, and at the same time analyzes the current situation and deficiencies of the system's supply resources, technologies and service layers. Third, this paper proposes an agricultural professional knowledge service system. The user interaction optimization plan starts from the human-computer interaction and interpersonal interaction dimensions of user interaction, analyzes and optimizes the resources, technologies and service layers of the agricultural knowledge system, realizes the friendly interaction of the system, improves the interaction incentive system, and builds a strong interactive knowledge chain community. [Results/Conclusions] The user interaction frame-work of the knowledge service system is designed, and based on this, we analyzed the current situation of user interaction in the agricultural knowledge system, and realized system optimization. The system can better stimulate user needs and understand user needs for the agricultural knowledge system, innovate functions, and provide high-quality personalized services, maintain the attractiveness and participation stickiness of users, benefit more user groups, and play a guiding role in the realization of system upgrades. Due to the lack of relevant knowledge of algorithm technology and lack of technical design for the optimization of agricultural professional knowledge service system, we need to explore the technical layer in the follow-up research; after the system optimization plan is proposed and implemented, experts and scholars need to further test its improvement effect, and propose construction to better realize the all-round optimization of the user interaction of the agricultural knowledge system.
Subjects: Other Disciplines >> Synthetic discipline submitted time 2023-03-31 Cooperative journals: 《农业图书情报学报》
Abstract： [Purpose/Significance] This paper investigates the research on author name disambiguation published in recent years, and reviews the development context of relevant research from the perspective of the impact of data on author name disambiguation methods, so as to provide reference for further research. [Method/Process] The papers related to author name disambiguation were collected from English research databases such as Web of Science, Scopus, Google Academic, ACM Digital Library, IEEE Xplore, ScienceDirect, Scopus and Springer Link, and Chinese research databases such as CNKI, CQVIP and WANFANG. The search results cover the relevant papers published from 1998 to 2021. On the premise of giving consideration to authority, influence and novelty, 46 publicationswere selected for review. There are many types and structures of author name disambiguation data. For example, literature feature information is generally presented in unstructured text, and the extracted features can be stored and represented in two-dimensional tables; Citation information and interpersonal relationship are network relational data, which can be stored and represented by graphs, key value pairs or two-dimensional tables. The fundamental reason for different data structures lies in their semantic differences, but the data structure itself determines its applicable algorithm. According to the structure of characteristic data used in the author name disambiguation task and the different corresponding data processing algorithms, the relevant research is divided into three categories: 1) disambiguation method based on literature characteristics, 2) disambiguation method based on social network and 3) disambiguation method by integrating external knowledge. The impact of data on the author name disambiguation method is examined from the data level. [Results/Conclusions] The analysis found that with the progress of technology, deep learning methods have been widely used. Compared with the improvement of the model, the feature learning and representation based on deep learning can significantly improve the effect of the author name disambiguation algorithm. In addition, in order to overcome the problem of insufficient data utilization by a single method and improve the utilization efficiency of data, the three methods show the trend of mutual combination and complementary gain. From the literature research results, there are few related studies on incremental author name disambiguation and multi-language author name disambiguation, which could be one of the directions for further research.
Subjects: Agriculture, Forestry,Livestock & Aquatic Products Science >> Other Disciplines of Agriculture, Forestry,Livestock & Aquatic Products Science submitted time 2023-02-17 Cooperative journals: 《智慧农业（中英文）》
Abstract： To further improve the simulation and estimation accuracy of the supply and demand process of agricultural products, a large number of agricultural data at the national and provincial levels since 1980 were used as the basic research sample, including production, planted area, food consumption, industrial consumption, feed consumption, seed consumption, import, export, price, GDP, population, urban population, rural population, weather and so on, by fully considering the impact factors of agricultural products such as varieties, time, income and economic development, a multi-agricultural products supply and demand forecasting model based on long short-term memory neural network (LSTM) was constructed in this study. The general thought of supply and demand forecasting model is packaging deep neural network training model as an I/O-opening modular model, reserving control interface for input of outside data, and realizing the indicators forecasting of supply and demand and matrixing of balance sheet. The input of model included forecasting balance sheet data of agricultural products, annual price data, general economic data, and international currency data since 2000. The output of model was balance sheet data of next decade since forecasting time. Under the premise of fully considering the mechanical constraints, the model used the advantages of deep learning algorithms in nonlinear model analysis and prediction to analyze and predict supply and demand of 9 main types of agricultural products, including rice, wheat, corn, soybean, pork, poultry, beef, mutton, and aquatic products. The production forecast results of 2019－2021 based on this model were compared and verified with the data published by the National Bureau of Statistics, and the mean absolute percentage error was 3.02%, which meant the average forecast accuracy rate of 2019－2021 was 96.98%. The average forecast accuracy rate was 96.10% in 2019, 98.26% in 2020, and 96.58% in 2021, which shows that with the increase of sample size, the prediction effect of intelligent learning model would gradually get better. The forecasting results indicate that the multi-agricultural supply and demand prediction model based on LSTM constructed in this study can effectively reflect the impact of changes in hidden indicators on the prediction results, avoiding the uncontrollable error introduced by manual experience intervention. The model can provide data production and technical support such as market warning, policy evaluation, resource management and public opinion analysis for agricultural production and management and macroeconomic regulation, and can provide intelligent technical support for multi-regional and inter-temporal agricultural outlook work by monitoring agricultural operation data in a timely manner.
Hits
Hits
Downloads
Comment