Search

The recommendation system is widely used in the digital economy to establish relationship between users and products. However, the data is usually scattered in different organizations in the form of "data islands" to protect privacy. To solve the above problem, we proposed an efficient and secure recommendation system based on federated matrix factorization, in which every user trains the model locally with their own data and uploads the calculated gradient vector instead of the original data. In the scheme, one server verifies the aggregation results transmitted by other servers and completes the model update without revealing user privacy. Additionally, users are allowed to drop out during the training process, and each participant's behavior will be recorded to calculate the rewards or punishments. Finally, we evaluated the performance from different perspectives and proved its feasibility. [ABSTRACT FROM AUTHOR]

In order to meet the demand for low-absorption sapphire crystal materials in important national defense projects in China, a self-produced crystal growth apparatus is developed in this study and the successful growth of low-absorption sapphire crystal can be achieved. The as-obtained crystal exhibits excellent transparency in the ultraviolet, visible, and near to midinfrared wavelength ranges. The transmittance is above 83% as the wavelength in the range of 250 nm to 400 nm, while the transmittance is above 85% as the wavelength in the range of 400 nm to 4 200 nm. On the basis of the characterization tests conducted on different regions of the crystal, the average dislocation density of crystal can be determined as 253. 19 cm-2. The rocking curve shows symmetrical and sharp peaks with a full width at half maximum of 14". In addition, it is worth noting that the light absorption coefficient of crystal is (23.3 -30.4) x 10-6 cm-1 at the wavelength of 1 064 nm, indicating that the impurity elements inside the crystal can be effectively controlled. Accordingly, the outstanding optical properties crystal can be applied in high-energy laser systems, space camera lenses and other relevant fields. [ABSTRACT FROM AUTHOR]

With the development of the Internet and the informatization construction of universities, the massive data accumulated by "campus big data" presents problems such as discreteness and sparseness. Students with abnormal behaviors have become an urgent problem to be solved in student behavior analysis. This paper proposes an early warning method for abnormal behaviour of college students based on multimodal fusion and an improved decision tree (EWMABCS-MFIDT). First, given the insufficient representation of student behavioral portraits and the problems of timeliness and dynamics in behavioral labels, a student behavioral portrait based on the multimodal fusion method is proposed. Second, aiming at the timeliness and backwardness of abnormal behavior prediction, based on student behavior classification prediction, this paper proposes an improved decision tree-based early warning method for abnormal student behavior. Finally, we design a student behavior analysis and early warning framework under the campus big data environment. Taking the abnormal early warning of students' academic performance as an example, compared with other early warning algorithms, the EWMABCS-MFIDT method can improve the accuracy of early warning and make students' educational work more targeted, personalized, and predictive. [ABSTRACT FROM AUTHOR]

Just-in-time (JIT) bug prediction is an effective quality assurance activity that identifies whether a code commit will introduce bugs into the mobile app, aiming to provide prompt feedback to practitioners for priority review. Since collecting sufficient labeled bug data is not always feasible for some mobile apps, one possible approach is to leverage cross-app models. In this work, we propose a new cross-triplet deep feature embedding method, called CDFE, for cross-app JIT bug prediction task. The CDFE method incorporates a state-of-the-art cross-triplet loss function into a deep neural network to learn high-level feature representation for the cross-app data. This loss function adapts to the cross-app feature learning task and aims to learn a new feature space to shorten the distance of commit instances with the same label and enlarge the distance of commit instances with different labels. In addition, this loss function assigns higher weights to losses caused by cross-app instance pairs than that by intra-app instance pairs, aiming to narrow the discrepancy of cross-app bug data. We evaluate our CDFE method on a benchmark bug dataset from 19 mobile apps with two effort-aware indicators. The experimental results on 342 cross-app pairs show that our proposed CDFE method performs better than 14 baseline methods. [ABSTRACT FROM AUTHOR]

Non‐functional requirements (NFRs) address important issues in software systems, and are vital in successful software. The NFR problems in a system reflect the complexity of that system. Ideally, NFRs are systematically investigated to determine the aspects that may be harmonious or cause conflicts so that conflicts should be detected as early as possible. The work aims to extend the scopes of the NFRs framework and softgoal interdependency graph (SIG) for modelling and analysing NFRs. A formal ontological approach to modelling NFR interactions is proposed, and a core NFR ontology is developed in the ontology web language. The mechanisms involved in NFR interactions with SIG models (through interdependencies in the softgoals) are investigated and formalised. The NFRs are first analyses using a semantic modelling process to provide evaluation criteria. Next, correlation rules for implicit softgoal interdependencies are proposed from knowledge of the domain, and rules for indirect semantic correlations are proposed by extending the delta evaluation method. Reasonable rules in the semantic web environment are defined to allow conflicts between NFRs in SIG instance models to be detected and analysed. A concepts proof is performed to illustrate tool support of the proposed method, and validate the rules for conflict detection. [ABSTRACT FROM AUTHOR]