Unpacking the innovation ecosystem construct: evolution, gaps and trends

Purpose The ecosystem perspective on innovation and business has emerged as the secret sauce of innovative organizations. While its theoretical foundations are premised on innovation system literature, the broad adoption of the ecosystem concept has resulted in conceptual ambiguity. The purpose of this paper is to tackle the ambiguous use of innovation ecosystem terminology and structure a conceptual frame for the field, identifying definitions of an innovation ecosystem and how the concept has been established in previous literature. Design/methodology/approach This paper examines the ambiguous use of terminology by reviewing the literature with bibliometric coupling and co-citation analysis by which thematic differences in ecosystem literature were identified. The study gathered the scientific publications from Thomson Reuters Web of Sciences Core Collection (n=4,681) from 1990 to 2015. Findings Six major bibliometrically coupled clusters were identified, of which the three largest clusters are innovation system studies, regional innovation studies and technological innovation studies. In addition, further analysis shows an emerging cluster that is focused on ecosystems, having its roots in eight seminal papers. This ecosystem research cluster includes seven sub-clusters, such as innovation ecosystem studies, business ecosystem studies and studies focusing on ecosystem development. Research limitations/implications The authors’ approach highlights how a lot of recent ecosystem studies actually belong to previous, well-developed research streams. However, there is also a separate, emergent research stream that includes the innovation and ecosystem studies. As a research implication, the paper concludes by suggesting the research agenda for further studies. Originality/value Even though literature on innovation systems and ecosystems is extensive literature, no studies have captured the emergence of the ecosystems approach and its relation with the systems of innovation literature.

A bibliometric review on innovation systems and ecosystems: a research agenda

Novel logic mining incorporating log linear approach

With the recent China electricity market reformation, an increasing portion of electricity is traded in wholesale electricity markets. However, collusion among generation companies reduces competition and raises market clearing prices. Based on the newly released electricity policies in China, a two-stage market clearing method is employed to mitigate collusion. In order to evaluate the effectiveness of this method, the system dynamics method is adopted. With consideration of the two aspects of collusion, i.e. capacity withholding and bidding at a high price, the causal feedback and quantitative relationship between the generator bidding process and the market clearing process are established. The sensitivities of the market parameters are analysed in a case study with real data collected from the Eastern China electricity market. The results can be used as a reference for policy makers.

Two-stage market clearing approach to mitigate generator collusion in Eastern China electricity market via system dynamics method

In the provided research, some of the Big Data most prospective usage domains connect with distinguished player groups found in the business ecosystem. Literature analysis is used to identify the state of the art of Big Data related research in the major domains of its use—namely, individual marketing, health treatment, work opportunities, financial services, and security enforcement. System theory was used to identify business ecosystem major player types disrupted by Big Data: individuals, small and mid-sized enterprises, large organizations, information providers, and regulators. Relationships between the domains and players were explained through new Big Data opportunities and threats and by players’ responsive strategies. System dynamics was used to visualize relationships in the provided model.

/pdf/big-data-for-business-ecosystem-players-noz6kem4dc.pdf

Big Data for Business Ecosystem Players

K-Nearest Neighbors (KNN) rule is a simple yet powerful classification technique in machine learning. Nevertheless, it suffers from some drawbacks such as high memory consumption, low time efficiency, class overlapping and difficulty of setting an appropriate K value. In this study, we propose an Improved K-Nearest Neighbor rule combining Prototype Selection and Local Feature Weighting (IKNN_PSLFW) to address the above issues in one framework. Differing from conventional prototype selection, IKNN_PSLFW not only selects the representative instances as prototypes, but also preserves the information of instances that are not selected. To deal with the class overlapping problem, IKNN_PSLFW explores the feature relevance in local regions by assigning different weights to different features. For an instance with unknown class label, IKNN_PSLFW uses three classification rules corresponding to three scenarios, according to the distance between the instance and each prototype, for classification. To evaluate the performance of IKNN_PSLFW, we conduct experimental study on 20 benchmark datasets. The experimental results show that compared with the conventional KNN rule, some state-of-the-art prototype selection methods and other machine learning algorithms, the proposed IKNN_PSLFW achieves promising classification performance with high time efficiency. 

K-nearest neighbors rule combining prototype selection and local feature weighting for classification

Forecasting the overnight return direction of stock market index combining global market indices: A multiple-branch deep learning approach

In recent years, we notice that the cooperation and competition among enterprises become much more complicated. Managers have to pay more attention to external cooperation from an ecological view. As e-business adoption becomes more pervasive, business ecosystems are shifting to e-business ecosystems. Because the situation is becoming more serious, in order to control the e-business ecosystem and earn profit from it, it is necessary for us to learn its structure and evolution. From the perspective of network science, this paper tries to connect complex network theories with e-business ecosystem research. We firstly analyse network structure of e-business ecosystem. Then an evolutionary model is proposed to describe the emergence and evolution of it. We finally use simulation and empirical methods to valid the theory we proposed.

/pdf/modelling-the-emergence-and-evolution-of-e-business-1lt46qiz3l.pdf

Modelling the Emergence and Evolution of e-Business Ecosystems from a Network Perspective

Integrating the sentiments of multiple news providers for stock market index movement prediction: A deep learning approach based on evidential reasoning rule

Strengthen the management of internet public opinion, and establishing the perfect public opinion supervision system has become a new important task around the world. The existing monitoring and counter mechanism of internet public opinion mainly includes public opinion monitoring, early-warning, response, neglecting the key link of evaluation, especially the evaluation of the effect of counter internet public opinion. Besides, the existing index is incomplete, unclear, or difficult to calculate. In this paper, based on the structure of internet public opinion ecology system, designing the model of ecological monitoring and response system. Relay on the decision-making layer, we get the 3 first-level indicators and giving detailed analysis then. We tend to guide the evaluation of the effect of dealing with the internet public opinion, and to improve the level of counter network public opinion.

Ruize Gao

Papers

K-nearest neighbors rule combining prototype selection and local feature weighting for classification

Forecasting the overnight return direction of stock market index combining global market indices: A multiple-branch deep learning approach

Modelling the Emergence and Evolution of e-Business Ecosystems from a Network Perspective

Integrating the sentiments of multiple news providers for stock market index movement prediction: A deep learning approach based on evidential reasoning rule

The Construction of Counter Effect Evaluation Index System of Internet Public Opinion Based on EMR