Hybrid Representation to Locate Vulnerable Lines of Code

Locating vulnerable lines of code in large software systems needs huge efforts from human experts. This explains the high costs in terms of budget and time needed to correct vulnerabilities. To minimize these costs, automatic solutions of vulnerabilities prediction have been proposed. Existing machine learning (ML)-based solutions face difficulties in predicting vulnerabilities in coarse granularity and in defining suitable code features that limit their effectiveness. To addressee these limitations, in the present work, the authors propose an improved ML-based approach using slice-based code representation and the technique of TF-IDF to automatically extract effective features. The obtained results showed that combining these two techniques with ML techniques allows building effective vulnerability prediction models (VPMs) that locate vulnerabilities in a finer granularity and with excellent performances (high precision (>98%), low FNR (<2%) and low FPR (<3%) which outperforms software metrics and are equivalent to the best performing recent deep learning-based approaches.

API Recommendation Based on WII-WMD#

Improving software development efficiency based on existing APIs is one of the hot researches in software engineering. Understanding and learning so many APIs in large software libraries is not easy and software developers prefer to provide only requirements descriptions to get the right API. In order to solve this problem, this paper proposes an API recommendation method based on WII-WMD, an improved similarity calculation algorithm. This method firstly structures the text, and then fully mines the semantic information in the text. Finally, it calculates the similarity between the user's query problem and the information described in the API document. The experiment results show that the API recommendation based on WII-WMD can improve the efficiency of the API recommendation system.

API Recommendation Based on WII-WMD

Improving software development efficiency based on existing APIs is one of the hot researches in software engineering. Understanding and learning so many APIs in large software libraries is not easy and software developers prefer to provide only requirements descriptions to get the right API. In order to solve this problem, this paper proposes an API recommendation method based on WII-WMD, an improved similarity calculation algorithm. This method firstly structures the text, and then fully mines the semantic information in the text. Finally, it calculates the similarity between the user's query problem and the information described in the API document. The experiment results show that the API recommendation based on WII-WMD can improve the efficiency of the API recommendation system.

Digital entrepreneurs in technology-based spinoffs: an analysis of hybrid value creation in retail public–private partnerships to tackle showrooming

Purpose Retail networks present new challenges in the business-to-business (B2B) collaboration between technology-based spinoffs and traditional businesses. This study aims to explore a public–private partnership (PPP) that leverages advanced digital technologies via spinoffs to tackle the key challenge of showrooming that retail shops are facing. Showrooming is the phenomenon in which shoppers go to the physical stores to gather in-depth product information, and later on, decide to buy the product from online retail competitors. Design/methodology/approach This research draws on a longitudinal qualitative study of a social context in which digital entrepreneurs are embedded. The empirical setting is a retail network in Italy, Germany and Finland with a particular focus on the process in which a PPP delivers innovation via spinoffs in the context of brick and mortar shops (B&M). The research design enables an understanding of the complexity of the phenomenon from a business and a social perspective. Findings New technology to tackle showrooming enables the creation of substantial hybrid value in retail partnerships. Spinoffs are key actors in leveraging digital technologies to create value faster and more tailored compared with large software companies. Spinoff entrepreneurs leverage on specific technologies (e.g. virtual reality and artificial intelligence) available inside organizations’ network (i.e. PPPs). Spinoffs are found to be a fundamental actor in the process of dealing with showrooming because of their time to market. Large software companies usually are not interested in approaching B&M shops because of the high operational costs of product customization for B&M shops. Practical implications Managers could use the success factors of the spinoffs in helping their B&M shops to improve both shopper experience and salesperson performance. For managers of B2B retail network, the results are useful towards increasing the involvement of shoppers while they are visiting physical stores, and it also improves salesperson performance. It also leads to the observation that cross-selling is one of the most effective responses to the phenomenon of showrooming. As practical implications for policymakers, the current research supports the view that PPPs should support the creation of spinoffs as a result of longitudinal innovation projects. Social implications Retail technologies leveraged from a PPP and commercialized by spinoffs are powerful tools to enable a better quality of salespeople’s life in the working place. At the same time, these new technologies help shop owners increase the retention rates, conversion rates and reduce short-term loss, increasing the likelihood of B&M shops to survive in the condition of extreme competition caused by the showrooming phenomenon. Originality/value This research proposes a model of hybrid value creation from networks in digital retail. The model indicates that PPPs create spinoffs to explore showrooming and deliver substantial hybrid value (i.e. business and social) for physical retail shops, mainly because it influences the companies’ growth, employee performance and customer satisfaction. This model expands the field of B2B marketing by identifying factors that enable spinoff creation from retail networks and proposes success factors and research propositions in retail networks.