What are the implications of big data in terms of big impacts? Our research focuses on the question, 'How are data analytics involved in policy analysis to create complementary values?' We address this from the perspective of bibliometrics. We initially investigate a set of articles published in Nature and Science, seeking cutting-edge knowledge to sharpen research hypotheses on what data science offers policy analysis. Based on a set of bibliometric models (e.g., topic analysis, scientific evolutionary pathways, and social network analysis), we follow up with studies addressing two aspects: (1) we examine the engagement of data science (including statistical, econometric, and computing approaches) in current policy analyses by analyzing articles published in top-level journals in the areas of political science and public administration; and (2) we examine the development of policy analysis-oriented data analytic models in top-level journals associated with computer science (including both artificial intelligence and information systems). Observations indicate that data science contribution to policy analysis is still an emerging area. Data scientists are moving further than policy analysts, due to technical difficulties in exploiting data analytic models. Integrating artificial intelligence with econometrics is identified as a particularly promising direction.

While there is a general recognition that breakthrough innovation is non-linear and requires an alignment between producers (supply) and users (demand), there is still a need for strategic intelligence about the emerging supply chains of new technological innovations. This technology delivery system (TDS) is an updated form of the TDS model and provides a promising chain-link approach to the supply side of innovation. Building on early research into supply-side TDS studies, we present a systematic approach to building a TDS model that includes four phases: (1) identifying the macroeconomic and policy environment, including market competition, financial investment, and industrial policy; (2) specifying the key public and private institutions; (3) addressing the core technical complements and their owners, then tracing their interactions through information linkages and technology transfers; and (4) depicting the market prospects and evaluating the potential profound influences on technological change and social developments. Our TDS methodology is illustrated using the field of Big Data & Analytics ("BDA").

Innovations around "Big Data" can be characterized in terms of rapid technology development and deployment dynamics. For this purpose, combining "tech mining" (extraction of usable intelligence) from publication and patent databases with tech mining of business-related databases can elucidate activities and interests of business communities regarding Big Data innovation pathways. In this paper, we focus on commercially oriented databases - ABI/INFORM as a source from which to extract business intents. We select the database to help gauge "hot topics" in industry with regard to Big Data. Our results show that certain types of firms can be clustered into thematic groups relating to Big Data discussions and activities. In the paper we demonstrate that such analyses can illuminate themes being pursued by businesses. Like social media analyses, this text mining can provide useful intelligence to inform more in-depth investigation mobilizing other data sources and techniques.