Search results

Any business that opts to adopt a recommender engine (RE) for various potential benefits must choose from the candidate solutions, by matching to the task of interest and domain. The purpose of this paper is to choose RE that fits best from a set of candidate solutions using rule-based automated machine learning (ML) approach. The objective is to draw trustworthy conclusion, which results in brand building, and establishing a reliable relation with customers and undeniably to grow the business.

An experimental quantitative research method was conducted in which the ML model was evaluated with diversified performance metrics and five RE algorithms by combining offline evaluation on historical and simulated movie data set, and the online evaluation on business-alike near-real-time data set to uncover the best-fitting RE.

The rule-based automated evaluation of RE has changed the testing landscape, with the removal of longer duration of manual testing and not being comprehensive. It leads to minimal manual effort with high-quality results and can possibly bring a new revolution in the testing practice to start a service line “Machine Learning Testing as a service” (MLTaaS) and the possibility of integrating with DevOps that can specifically help agile team to ship a fail-safe RE evaluation product targeting SaaS (software as a service) or cloud deployment.

A small data set was considered for A/B phase study and was captured for ten movies from three theaters operating in a single location in India, and simulation phase study was captured for two movies from three theaters operating from the same location in India. The research was limited to Bollywood and Ollywood movies for A/B phase, and Ollywood movies for simulation phase.

The best-fitting RE facilitates the business to make personalized recommendations, long-term customer loyalty forecasting, predicting the company's future performance, introducing customers to new products/services and shaping customer's future preferences and behaviors.

The proposed rule-based ML approach named “2-stage locking evaluation” is self-learned, automated by design and largely produces time-bound conclusive result and improved decision-making process. It is the first of a kind to examine the business domain and task of interest. In each stage of the evaluation, low-performer REs are excluded which leads to time-optimized and cost-optimized solution. Additionally, the combination of offline and online evaluation methods offer benefits, such as improved quality with self-learning algorithm, faster time to decision-making by significantly reducing manual efforts with end-to-end test coverage, cognitive aiding for early feedback and unattended evaluation and traceability by identifying the missing test metrics coverage.

To better understand the impact of choosing a carbon data provider for the estimated portfolio emissions across four asset classes. This is important, as prior literature has suggested that Environmental, Social and Governance scores across providers have low correlation.

The authors compare carbon data from four data providers for developed and emerging equity markets and investment grade and high-yield corporate bond markets.

Data on scope 1 and scope 2 is similar across the four data providers, but for scope 3 differences can be substantial. Carbon emissions data has become more consistent across providers over time.

The authors examine the impact of different carbon data providers at the asset class level. Portfolios that invest only in a subset of the asset class may be affected differently. Because “true” carbon emissions are not known, the authors cannot investigate which provider has the most accurate carbon data.

The impact of choosing a carbon data provider is limited for scope 1 and scope 2 data for equity markets. Differences are larger for corporate bonds and scope 3 emissions.

The authors compare carbon accounting metrics on scopes 1, 2 and 3 of corporate greenhouse gas emissions carbon data from multiple providers for developed and emerging equity and investment grade and high yield investment portfolios. Moreover, the authors show the impact of filling missing data points, which is especially relevant for corporate bond markets, where data coverage tends to be lower.

Sensing coverage is a critical issue in sensor network deployments. The paper aims to propose a novel scheme to maintain the sensing coverage in sensor networks, which is termed coverage‐aware self‐scheduling (CASS).

The paper describes a generic unifying framework to incorporate different connectivity and coverage maintenance schemes. Simulations are carried out under the framework by integrating CASS with an existing connectivity maintenance scheme ‐ the low‐energy adaptive clustering hierarchy.

Different from the existing work on coverage maintenance, CASS probabilistically schedules sensing activities according to the sensor's contribution to the sensing coverage of the whole sensor network. CASS reduces the number of active sensors to maintain certain coverage. Besides the sensing coverage, the connectivity of the network topologies is required for the purpose of communicating among sensors to collect sensing data. Simulation results show that CASS can considerably improve the energy efficiency of sensing coverage with low communication and computation overhead.

The paper shows that CASS is designed to allow sensors with higher coverage contribution to have more chance to sense.

The purpose of this paper is to present a verification methodology for custom micro coded components designed for Avionics projects. Every electronic hardware which will be developed for an aircraft must be designed with the compliance of DO-254 processes. Requirements are the key elements of the aviation. All the requirements must be covered by the design to be considered as completed. Therefore, verification of the custom micro coded components against requirements should be comprehensively addressed. The verification using the manual testing approach is less preferable, as humans can possibly make mistakes. Therefore, the most used verification method today is the automated simulation.

The industry has developed a common methodology for generating automated testbenches by following the standardized guideline. This methodology is named as the universal verification methodology (UVM). In this paper, the verification study of ARINC-429 data bus digital design is presented to describe the DO-254 verification process using the UVM.

The results are supported with functional coverage and code coverage in addition to the assertions. It is observed that the design worked correctly.

To the best of the authors’ knowledge, this is the first study comprehensively describing the DO-254 verification process and demonstrating it by the UVM application of ARINC-429 on programmable logic devices.

The purpose of this paper is to present a new multi‐objective clonal selection algorithm (MCSA) for the solution of electromagnetic optimization problems.

The method performs the somatic hypermutation step using different probability distributions, balancing the local search in the algorithm. Furthermore, it includes a receptor editing operator that implicitly realizes a dynamic search over the landscape.

In order to illustrate the efficiency of MCSA, its performance is compared with the nondominated sorting genetic algorithm II (NSGA‐II) in some analytical problems and in the well‐known TEAM benchmark Problem 22. Three performance evaluation techniques are used in the comparison, and the effect of each operator of the MCSA in its accomplishment is estimated.

In the analytical problems, the MCSA enhanced both the extension and uniformity in its solutions, providing better Pareto‐optimal sets than the NSGA‐II. In the Problem 22, the MCSA also outperformed the NSGA‐II. The MCSA was not dominated by the NSGA‐II in the three variables case and clearly presented a better convergence speed in the eight variables problem.

This paper could be useful for researchers who deal with multi‐objective optimization problems involving high‐computational cost.

The new operators incorporated in the MCSA improved both the extension, uniformity and the convergence speed of the solutions, in terms of the number of function evaluations, representing a robust tool for real‐world optimization problems.

A recommendation algorithm is typically applied to speculate on users’ preferences based on their behavioral characteristics. The purpose of this paper is to provide a systematic review of recommendation systems by collecting related journal articles from the last five years (i.e. from 2014 to 2018). This paper aims to study the correlations between recommendation technologies and e-learning systems.

The paper reviews the relevant articles using five assessment aspects. A coding scheme was put forward that includes the following: the metrics for the e-learning system, the evaluation metrics for the recommendation algorithms, the recommendation filtering technology, the phases of the recommendation process and the learning outcomes of the system.

The research indicates that most e-learning systems will adopt the adaptive mechanism as a primary metric, and accuracy is a vital evaluation indicator for recommendation algorithms. In existing e-learning recommender systems, the most common recommendation filtering technology is hybrid filtering. The information collection phase is an important process recognized by most studies. Finally, the learning outcomes of the recommender system can be achieved through two key indicators: affections and correlations.

The recommendation technology works effectively in closing the gap between the information producer and the information consumer. This technology could help learners find the information they are interested in as well as send them a valuable message. The opportunities and challenges of the current study are discussed; the results of this study could provide a guideline for future research.

The purpose of this research is to study the Multi-Skill Resource-Constrained Multi-Project Scheduling Problem (MSRCMPSP), where (1) durations of activities depend on the familiarity levels of assigned workers, (2) more efficient workers demand higher per-day salaries, (3) projects have different due dates and (4) the budget of each period varies over time. The proposed model is bi-objective, and its objectives are minimization of completion times and costs of all projects, simultaneously.

This paper proposes a two-phase approach based on the Statistical Process Control (SPC) to solve this problem. This approach aims to develop a control chart so as to monitor the performance of an optimizer during the optimization process. In the first phase, a multi-objective statistical model has been used to obtain control limits of this chart. To solve this model, a Multi-Objective Greedy Randomized Adaptive Search Procedure (MOGRASP) has been hired. In the second phase, the MSRCMPSP is solved via a New Version of the Multi-Objective Variable Neighborhood Search Algorithm (NV-MOVNS). In each iteration, the developed control chart monitors the performance of the NV-MOVNS to obtain proper solutions. When the control chart warns about an out-of control state, a new procedure based on the Conway’s Game of Life, which is a cellular automaton, is used to bring the algorithm back to the in-control state.

The proposed two-phase approach has been used in solving several standard test problems available in the literature. The results are compared with the outputs of some other methods to assess the efficiency of this approach. Comparisons imply the high efficiency of the proposed approach in solving test problems with different sizes.

The proposed model and approach have been used to schedule multiple projects of a construction company in Iran. The outputs show that both the model and the NV-MOVNS can be used in real-world multi-project scheduling problems.

Due to the numerous numbers of studies reviewed in this research, the authors discovered that there are few researches on the multi-skill resource-constrained multi-project scheduling problem (MSRCMPSP) with the aforementioned characteristics. Moreover, none of the previous researches proposed an SPC-based solution approach for meta-heuristics in order to solve the MSRCMPSP.

In recent years, governmental regulations and the pressure of non-governmental organizations have convinced corporations to consider sustainable issues in their decisions. A simultaneous design of forward and reverse logistics can keep us away from sub-optimality caused by tackling these two phases (forward and reverse logistics) separately.

Hence, this paper presents a new multi-objective mathematical model for integrated forward and reverse logistics regarding economic, environmental and social issues. A new hybrid multi-objective metaheuristic algorithm is developed to obtain a set of efficient solutions (Pareto solutions). The proposed algorithm hybridizes a well-known, non-dominated genetic algorithm (NSGA-II) with a simulated annealing algorithm.

To validate the algorithm, its results are compared to the obtained solutions from simple NSGA-II with respect to some comparison metrics. The numerical results show the efficiency of the proposed algorithm. Finally, concluding remarks and future research directions are provided.

By applying a model presented in this paper, one can reach to sustainable and integrated logistics network which considers forward and reverse flow of commodities simultaneously.

The purpose of this study is to contribute to the understanding of how the potential of altmetrics varies around the world by measuring the percentage of articles with non-zero metrics (coverage) for articles published from a developing region (Latin America).

This study uses article metadata from a prominent Latin American journal portal, SciELO, and combines it with altmetrics data from Altmetric.com and with data collected by author-written scripts. The study is primarily descriptive, focusing on coverage levels disaggregated by year, country, subject area, and language.

Coverage levels for most of the social media sources studied was zero or negligible. Only three metrics had coverage levels above 2 per cent – Mendeley, Twitter, and Facebook. Of these, Twitter showed the most significant differences with previous studies. Mendeley coverage levels reach those found by previous studies, but it takes up to two years longer for articles to be saved in the reference manager. For the most recent year, coverage was less than half than what was found in previous studies. The coverage levels of Facebook appear similar (around 3 per cent) to that of previous studies.

The Altmetric.com data used for some of the analyses were collected for a six month period. For other analyses, Altmetric.com data were only available for a single country (Brazil).

The results of this study have implications for the altmetrics research community and for any stakeholders interested in using altmetrics for evaluation. It suggests the need of careful sample selection when wishing to make generalizable claims about altmetrics.

This paper aims to improve the recommendations performance for cold-start users and controversial items. Collaborative filtering (CF) generates recommendations on the basis of similarity between users. It uses the opinions of similar users to generate the recommendation for an active user. As a similarity model or a neighbor selection function is the key element for effectiveness of CF, many variations of CF are proposed. However, these methods are not very effective, especially for users who provide few ratings (i.e. cold-start users).

A new user similarity model is proposed that focuses on improving recommendations performance for cold-start users and controversial items. To show the validity of the authors’ similarity model, they conducted some experiments and showed the effectiveness of this model in calculating similarity values between users even when only few ratings are available. In addition, the authors applied their user similarity model to a recommender system and analyzed its results.

Experiments on two real-world data sets are implemented and compared with some other CF techniques. The results show that the authors’ approach outperforms previous CF techniques in coverage metric while preserves accuracy for cold-start users and controversial items.

In the proposed approach, the conditions in which CF is unable to generate accurate recommendations are addressed. These conditions affect CF performance adversely, especially in the cold-start users’ condition. The authors show that their similarity model overcomes CF weaknesses effectively and improve its performance even in the cold users’ condition.