Web App Research Articles

Predicting Lewis Acidity: Machine Learning the Fluoride Ion Affinity of p‐Block‐Atom‐Based Molecules

Abstract“How strong is this Lewis acid?” is a question researchers often approach by calculating its fluoride ion affinity (FIA) with quantum chemistry. Here, we present FIA49k, an extensive FIA dataset with 48,986 data points calculated at the RI‐DSD‐BLYP‐D3(BJ)/def2‐QZVPP//PBEh‐3c level of theory, including 13 different p‐block atoms as the fluoride accepting site. The FIA49k dataset was used to train FIA‐GNN, two message‐passing graph neural networks, which predict gas and solution phase FIA values of molecules excluded from training with a mean absolute error of 14 kJ mol−1 (r2=0.93) from the SMILES string of the Lewis acid as the only input. The level of accuracy is notable, given the wide energetic range of 750 kJ mol−1 spanned by FIA49k. The model's value was demonstrated with four case studies, including predictions for molecules extracted from the Cambridge Structural Database and by reproducing results from catalysis research available in the literature. Weaknesses of the model are evaluated and interpreted chemically. FIA‐GNN and the FIA49k dataset can be reached via a free web app (www.grebgroup.de/fia‐gnn).

Predicting Lewis Acidity: Machine Learning the Fluoride Ion Affinity of p-Block-Atom-Based Molecules.

"How strong is this Lewis acid?" is a question researchers often approach by calculating its fluoride ion affinity (FIA) with quantum chemistry. Here, we present FIA49k, an extensive FIA dataset with 48,986 data points calculated at the RI-DSD-BLYP-D3(BJ)/def2-QZVPP//PBEh-3c level of theory, including 13 different p-block atoms as the fluoride accepting site. The FIA49k dataset was used to train FIA-GNN, two message-passing graph neural networks, which predict gas and solution phase FIA values of molecules excluded from training with a mean absolute error of 14 kJ mol-1 (r2=0.93) from the SMILES string of the Lewis acid as the only input. The level of accuracy is notable, given the wide energetic range of 750 kJ mol-1 spanned by FIA49k. The model's value was demonstrated with four case studies, including predictions for molecules extracted from the Cambridge Structural Database and by reproducing results from catalysis research available in the literature. Weaknesses of the model are evaluated and interpreted chemically. FIA-GNN and the FIA49k dataset can be reached via a free web app (www.grebgroup.de/fia-gnn).

Predictive Model in Production through Progressive Web Applications to Forecast Moniliasis in Cacao.

Cocoa is considered a significant crop in Ecuador, as it represents a favorable source of income for the country's economy, thanks to the remarkable quality of the product. However, it faces a significant issue in its crops: moniliasis, a fungal disease that attacks cocoa cultivation, is present in most Latin American countries. Consequently, this leads to decreased cocoa production and a lower final product quality. The study focuses on designing a predictive production model through a progressive web application to forecast moniliasis in cocoa. The objective is to create an application that anticipates the presence of this disease, thereby contributing to the improvement of the local economy for all farmers. Various methodologies were employed, including bibliographic methods, design science research methodology, and machine learning models. The results obtained from this research indicate that the Gradient Boosting Classifier is the algorithm that best fits the provided dataset. Once this algorithm was identified, a progressive web application was developed and made available for public use by farmers. Furthermore, the efficiency of the predictive model was verified using the statistical method of central tendency, demonstrating that the predictive model is beneficial, primarily by saving farmers a significant amount of time. Anticipating the disease enables timely preventive and corrective measures, which could reduce losses in cocoa production and enhance the quality of the final product. Keywords: cacao, moniliasis, predictive model, progressive web apps, supervised learning

Predictive Model in Production through Progressive Web Applications to Forecast Moniliasis in Cacao.

Cocoa is considered a significant crop in Ecuador, as it represents a favorable source of income for the country's economy, thanks to the remarkable quality of the product. However, it faces a significant issue in its crops: moniliasis, a fungal disease that attacks cocoa cultivation, is present in most Latin American countries. Consequently, this leads to decreased cocoa production and a lower final product quality. The study focuses on designing a predictive production model through a progressive web application to forecast moniliasis in cocoa. The objective is to create an application that anticipates the presence of this disease, thereby contributing to the improvement of the local economy for all farmers. Various methodologies were employed, including bibliographic methods, design science research methodology, and machine learning models. The results obtained from this research indicate that the Gradient Boosting Classifier is the algorithm that best fits the provided dataset. Once this algorithm was identified, a progressive web application was developed and made available for public use by farmers. Furthermore, the efficiency of the predictive model was verified using the statistical method of central tendency, demonstrating that the predictive model is beneficial, primarily by saving farmers a significant amount of time. Anticipating the disease enables timely preventive and corrective measures, which could reduce losses in cocoa production and enhance the quality of the final product. Keywords: cacao, moniliasis, predictive model, progressive web apps, supervised learning.

Fighting moniliasis in Orellana with sensors and PWA for sustainable agriculture

The primary objective of this research was to enhance cocoa production and quality in tropical countries, such as Latin America and Africa, where cocoa cultivation plays a pivotal role in the economy of rural communities. The primary challenge addressed in this study was moniliasis, a fungal disease that affects cocoa fruits and leads to a significant decline in crop production and quality. A multidisciplinary approach was employed to tackle this issue, combining sensors, MongoDB Compass databases, Progressive Web Applications (PWAs), and predictive models. A research methodology incorporating predictive analysis techniques, particularly the logistic regression method, was utilized to achieve early detection and efficient management of moniliasis. Data collection instruments included sensors monitoring vital environmental factors like humidity and temperature alongside MongoDB Compass databases for storing and managing the gathered data. Furthermore, a PWA was developed for real-time data collection and analysis. The results of implementing this sensor-based tool in cocoa cultivation were highly effective. Early detection of moniliasis allowed for more precise preventive and corrective measures, resulting in a significant improvement in cocoa production and quality. These results were substantiated with concrete data demonstrating the tool's efficacy. Keywords. Data Prediction: Models and Applications; Efficient MongoDB Database Management; High-Quality Cocoa; Moniliasis: Treatment and Prevention; Progressive Web Apps (PWA): User Experience.

Fighting moniliasis in Orellana with sensors and PWA for sustainable agriculture

The primary objective of this research was to enhance cocoa production and quality in tropical countries, such as Latin America and Africa, where cocoa cultivation plays a pivotal role in the economy of rural communities. The primary challenge addressed in this study was moniliasis, a fungal disease that affects cocoa fruits and leads to a significant decline in crop production and quality. A multidisciplinary approach was employed to tackle this issue, combining sensors, MongoDB Compass databases, Progressive Web Applications (PWAs), and predictive models. A research methodology incorporating predictive analysis techniques, particularly the logistic regression method, was utilized to achieve early detection and efficient management of moniliasis. Data collection instruments included sensors monitoring vital environmental factors like humidity and temperature alongside MongoDB Compass databases for storing and managing the gathered data. Furthermore, a PWA was developed for real-time data collection and analysis. The results of implementing this sensor-based tool in cocoa cultivation were highly effective. Early detection of moniliasis allowed for more precise preventive and corrective measures, resulting in a significant improvement in cocoa production and quality. These results were substantiated with concrete data demonstrating the tool's efficacy. Keywords: Data Prediction: Models and Applications; Efficient MongoDB Database Management; HighQuality Cocoa; Moniliasis: Treatment and Prevention; Progressive Web Apps (PWA): User Experience.

Easy-Amanida: An R Shiny application for the meta-analysis of aggregate results in clinical metabolomics using Amanida and Webchem.

Meta-analysis is a useful tool in clinical research, as it combines the results of multiple clinical studies to improve precision when answering a particular scientific question. While there has been a substantial increase in publications using meta-analysis in various clinical research topics, the number of published meta-analyses in metabolomics is significantly lower compared to other omics disciplines. Metabolomics is the study of small chemical compounds in living organisms, which provides important insights into an organism's phenotype. However, the wide variety of compounds and the different experimental methods used in metabolomics make it challenging to perform a thorough meta-analysis. Additionally, there is a lack of consensus on reporting statistical estimates, and the high number of compound naming synonyms further complicates the process. Easy-Amanida is a new tool that combines two R packages, "amanida" and "webchem", to enable meta-analysis of aggregate statistical data, like p-value and fold-change, while ensuring the compounds naming harmonization. The Easy-Amanida app is implemented in Shiny, an R package add-on for interactive web apps, and provides a workflow to optimize the naming combination. This article describes all the steps to perform the meta-analysis using Easy-Amanida, including an illustrative example for interpreting the results. The use of aggregate statistics metrics extends the use of Easy-Amanida beyond the metabolomics field.

Designing an App to Support Measurement-Based Peer Supervision of Frontline Health Workers Delivering Brief Psychosocial Interventions in Texas: Multimethod Study.

The unmet need for mental health care affects millions of Americans. A growing body of evidence in implementation science supports the effectiveness of task sharing in the delivery of brief psychosocial interventions. The digitization of training and processes supporting supervision can rapidly scale up task-shared interventions and enable frontline health workers (FLWs) to learn, master, and deliver interventions with quality and support. We aimed to assess the perceived feasibility and acceptability of a novel mobile and web app designed and adapted to support the supervision, training, and quality assurance of FLWs delivering brief psychosocial interventions. We followed human-centered design principles to adapt a prototype app for FLWs delivering brief psychosocial interventions for depression, drawing from an app previously designed for use in rural India. Using a multimethod approach, we conducted focus group sessions comprising usability testing and group interviews with FLWs recruited from a large health system in Texas to assess the feasibility and acceptability of the app. The positive System Usability Scale was used to determine the app's overall usability. We also calculated the participants' likelihood of recommending the app to others using ratings of 0 to 10 from least to most likely (net promoter score). Focus group transcripts were coded and analyzed thematically, and recommendations were summarized across 4 key domains. A total of 18 FLWs varying in role and experience with client care participated in the study. Participants found the app to be usable, with an average System Usability Scale score of 72.5 (SD 18.1), consistent with the industry benchmark of 68. Participants' likelihood of recommending the app ranged from 5 to 10, yielding a net promoter score of 0, indicating medium acceptability. Overall impressions of the app from participants were positive. Most participants (15/18, 83%) found the app easy to access and navigate. The app was considered important to support FLWs in delivering high-quality mental health care services. Participants felt that the app could provide more structure to FLW training and supervision processes through the systematic collection and facilitation of performance-related feedback. Key concerns included privacy-related and time constraints regarding implementing a separate peer supervision mechanism that may add to FLWs' workloads. We designed, built, and tested a usable, functional mobile and web app prototype that supports FLW-delivered psychosocial interventions in the United States through a structured supervision mechanism and systematic collection and review of performance measures. The app has the potential to scale the work of FLWs tasked with delivering these interventions to the hardest-to-reach communities they serve. The results of this project will inform future work to evaluate the app's use and efficacy in real-world settings to support task-shared mental health programs across the United States.

A Novel and Extensible Remote Sensing Collaboration Platform: Architecture Design and Prototype Implementation

Geospatial data, especially remote sensing (RS) data, are of significant importance for public services and production activities. Expertise is critical in processing raw data, generating geospatial information, and acquiring domain knowledge and other remote sensing applications. However, existing geospatial service platforms are more oriented towards the professional users in the implementation process and final application. Building appropriate geographic applications for non-professionals remains a challenge. In this study, a geospatial data service architecture is designed that links desktop geographic information system (GIS) software and cloud-based platforms to construct an efficient user collaboration platform. Based on the scalability of the platform, four web apps with different themes are developed. Data in the fields of ecology, oceanography, and geology are uploaded to the platform by the users. In this pilot phase, the gap between non-specialized users and experts is successfully bridged, demonstrating the platform’s powerful interactivity and visualization. The paper finally evaluates the capability of building spatial data infrastructures (SDI) based on GeoNode and discusses the current limitations. The support for three-dimensional data, the improvement of metadata creation and management, and the fostering of an open geo-community are the next steps.

NSDIE: Noise Suppressing Dark Image Enhancement Using Multiscale Retinex and Low-Rank Minimization

It is inevitable for dark images to have crucial information obscured by low-light conditions, which are worsened by the presence of noise in these images. This work introduces a groundbreaking solution, Noise-Suppressing Dark Image Enhancement for Web Apps (NSDIE), to address the challenging task of enhancing low-light images marred by noise. The proposed work utilizes a low-rank model with simultaneous enhancement of reflectance and illumination components to improve the nighttime scenes while also eradicating the present noise of the image. The reflectance component is further processed using a multiscale retinex model to compensate for the possible color distortions while the illumination component is enhanced using the camera response model to ensure the genuineness of the scene. The proposed work is also tested for a standalone application and is presented to the user through a web portal to aid the concerns of dark image enhancement in the daily life of the user. Rigorous quantitative and qualitative analyses assert NSDIE's superiority over existing techniques, establishing its pivotal role in addressing the critical concern of dark image enhancement.

Efficacy of a facial-aging web app on sun protection behaviors among primary school students in Iran: a randomized controlled trial

BackgroundSkin cancers resulting from excessive exposure to solar ultraviolet (UV) radiation are on the rise. This study aims to investigate the impact of facial-aging app intervention on promoting safe and healthy behaviors and its influence on reducing students' UV exposure.MethodUtilizing a Pretest–Posttest repeated-measures design, we developed a theory-guided web app on the WhatsApp platform, named the Sunshine and Skin Health app. This app allows users to visualize their altered faces in three stages of adolescence, middle age, and old age based on sun protection behavior. The intervention continued within WhatsApp, incorporating 27 health messages grounded in the PMT theory, eight educational files, and a skin cancer video clip. The primary outcome is the change in sun protection behavior between the two groups (intervention and control) immediately after the intervention (T2) and the secondary outcome is the change in sun protection behavior between the two groups at 3 months follow-up (T3). The data are analyzed in SPSS 22 and a significance level of 0.05 is considered.ResultsThe results revealed no significant difference between the two groups before the intervention. However, in the intervention group, there were significant differences in the utilization of sunglasses, hats, and sunscreen in the last month, as well as sunscreen reapplication after washing their hands and face, both immediately after the intervention and at the 3-month follow-up, compared to the control group (P = 0.001). Furthermore, a significant intervention effect, time effect, and interaction effect between group and time were observed in behaviors related to using sunscreen in the last month and sunscreen reapplication after washing hands and face (P = 0.001). Specifically, the intervention group exhibited a significant difference from Time 1 to 2 and from Time 1 to 3 (p = 0.001), but no significant difference from Time 2 to 3. In contrast, the control group did not show any significant differences over time.ConclusionsThis study indicated that the Facial-Aging web app can effectively encourage safe behaviors in sunlight. To ensure the maintenance and sustainability of these behaviors over the long term, it is crucial to consider implementing booster sessions.Trial registrationIranian Registry of Clinical Trials IRCT20200924048825N1. Registered prospectively on 8 February 2021.

VehiWalnut: Smart Fuel Management with Real-Time Data and User Interaction

This research paper examines the creation and execution of VehiWalnut, a cutting-edge system developed to enhance fuel management processes. VehiWalnut incorporates cutting-edge technologies such as data visualization, cloud-based chatbot evaluation, and deep learning to improve efficiency and user satisfaction. A thorough literature review is conducted to identify important studies in data visualization, Web API analysis, chatbot performance evaluation, and deep learning techniques for machine learning chatbots. These studies are then combined to form the basis of the research. The text describes the development of VehiWalnut, emphasizing its diverse functionalities and the use of the Python programming language for its implementation. The system allows users to enter fuel fill-up data through WhatsApp messages, which are processed using regular expressions to extract and insert the data into a SQLite database. VehiWalnut allows users to request detailed reports on fuel expenses, timing of fuel refills, predictions for the next refill date, and other related information. The system’s effectiveness is assessed by conducting case studies and performance evaluations, showcasing its capacity to enhance fuel management procedures for individual and commercial users. The research findings offer valuable insights into incorporating advanced technologies to improve fuel management systems, thereby paving the way for future advancements in this domain.

New Methodology to Evaluate and Optimize Indoor Ventilation Based on Rapid Response Sensors.

The recent pandemic increased attention to the need for appropriated ventilation and good air quality as efficient measures to achieve safe and healthy indoor air. This work provides a novel methodology for continuously evaluating ventilation in public areas using modern rapid response sensors (RRS). This methodology innovatively assesses the ventilation of a space by combining a quantitative estimation of the real air exchange in the space-obtained from CO2 experimental RRS measurements and the characteristics of and activity in the space-and indoor and outdoor RRS measurements of other pollutants, with healthy recommendations from different organisations. The methodology allows space managers to easily evaluate, in a continuous form, the appropriateness of their ventilation strategy, thanks to modern RRS measurements and direct calculations (implemented here in a web app), even in situations of full activity. The methodology improves on the existing standards, which imply the release of tracer gases and expert intervention, and could also be used to set a control system that measures continuously and adapts the ventilation to changes in indoor occupancy and activity, guaranteeing safe and healthy air in an energy-efficient way. Sample public concurrence spaces with different conditions are used to illustrate the methodology.

Developing an Educational Chatbot for Scientific Data Management Course Using DialogFlow

The COVID-19 pandemic has radically altered the worldwide learning environments, setting the stage for Electronic Learning (E-Learning) advancement, where remote learning is facilitated through digital tools. The key stakeholders (professors, teaching assistants, and students) face bottlenecks as they shift to online education providing blended learning. New learning tools based on Natural Language Processing (NLP) are provided. Designing a chatbot is one of the solutions to deal with this issue. Chatbots are simple computer programs that attempt to simulate human conversation using Artificial Intelligence (AI) and NLP. They allow learners to have a standardized learning environment. In this paper, we have set up a chatbot application named Scientific Data Management BOT (SDMBOT) for handling E-Learning activities specifically for Scientific Data Management (SDM) courses based on AI and NLP techniques using Dialogflow Framework, a Google development platform for building NLP-based human-computer interface solutions. SDMBOT was trained on a dataset that was specifically created based on course content. The web or mobile app, via which our built chatbot is available, is used for student interaction. Students can ask the chatbot questions concerning SDM, and the chatbot will process the message and respond to the user by displaying the proper result. The accuracy of the SDM chatbot, which is calculated by using a confusion matrix indicated that our chatbot is 74 % accurate.

Potential Economic and Ecosystem Performances of Some Mediterranean Fruit Plants in an Urban Context

This study focuses on the possibility of using fruit tree species, from an urban area in central Italy, to evaluate their possible productive, ecosystem, and economic contribution. The realization of the food forest was conducted through the use of a web app that can evaluate the climatic and ecosystem performance of trees in the city. This simulation can provide the city’s government with useful information on the ecosystem and the socioeconomic benefits of planting fruit trees in an urban park. Among the four chosen species, both Ficus carica and Morus nigra showed good potential fruit production of 1300 kg and 865 kg and a CO2 storage of 2.5 and 1.5 tons. The production and economic potentials from selling the fruits of the selected species were evaluated, and the Morus nigra species showed the highest economic value over a 50-year period of about 6000 USD, with an average price of 7 USD/kg. Another positive aspect of edible food forest landscapes is their ability to give rise to an ecosystem and habitat that can attract animals, birds, and wildlife, consequently improving urban ecosystem biodiversity.

Implementing and testing a U-space system: lessons learnt

Within the framework of the European Union’s Horizon 2020 research and innovation program, one of the main goals of the Labyrinth project, which ended on May 2023, was to develop and test the Conflict Management services of a U-space-based Unmanned Traffic Management (UTM) system. The U-space ConOps provides a high-level description of the architecture, requirements and functionalities of these systems, but the implementer has a certain degree of freedom in aspects like the techniques used or some policies and procedures. The current document describes some of those implementation decisions. The prototype included, at least in a basic version, part of the services defined by the ConOps, namely e-identification, Tracking, Geo-awareness, Drone Aeronautical Information Management, Geo-fence Provision, Operation Plan Preparation/Optimization, Operation Plan Processing, Strategic and Tactical Conflict Resolution, Emergency Management, Monitoring, Traffic Information and Legal Recording. Besides, a web app interface was developed for the operator/pilot. The system was tested in simulations and real visual line of sight (VLOS) and beyond VLOS (BVLOS) flights, with both vertical take-off and landing (VTOL) and fixed-wing platforms, while assisting final users interested in incorporating drones to support their daily tasks. The three-year experience developing and testing the environment provided many lessons at different levels: functionalities, compatibility, procedures, information, usability, ground control station (GCS) integration and aircrew roles during the mission.

PC 환경에서의 OTT 콘텐츠 데이터 분석 방법

Due to technological advancements in viewing devices and the COVID-19 pandemic, a lot of OTT-only content is being produced and distributed as people shift from traditional movie theater viewing and broadcasters' fixed TV viewing to free-form OTT viewing using wired and wireless internet. As a result, the ability to leverage data from OTT audiences has become critical to the competitiveness of the industry. However, third parties other than OTT content providers are facing difficulties in acquiring OTT viewer data. In this paper, as a way to overcome the shortcomings of existing viewer data acquisition, we propose a method to extract audio and video data by developing an OTT viewing data acquisition agent using Web APIs by adopting a web browser environment that does not affect the performance of OS and viewing devices, so that third-party companies that need viewing data can utilize it.

Digital Gamification Tools to Enhance Vaccine Uptake: Scoping Review.

Gamification has been used successfully to promote various desired health behaviors. Previous studies have used gamification to achieve desired health behaviors or facilitate their learning about health. In this scoping review, we aimed to describe digital gamified tools that have been implemented or evaluated across various populations to encourage vaccination, as well as any reported effects of identified tools. We searched Medline, Embase, CINAHL, the Web of Science Core Collection, the Cochrane Database of Systematic Reviews, the Cochrane Central Register of Controlled Trials, Academic Search Premier, PsycInfo, Global Health, and ERIC for peer-reviewed papers describing digital gamified tools with or without evaluations. We also conducted web searches with Google to identify digital gamified tools lacking associated publications. We consulted 12 experts in the field of gamification and health behavior to identify any papers or tools we might have missed. We extracted data about the target population of the tools, the interventions themselves (eg, type of digital gamified tool platform, type of disease/vaccine, type and design of study), and any effects of evaluated tools, and we synthesized data narratively. Of 1402 records, we included 28 (2%) peer-reviewed papers and 10 digital gamified tools lacking associated publications. The experts added 1 digital gamified tool that met the inclusion criteria. Our final data set therefore included 28 peer-reviewed papers and 11 digital gamified tools. Of the 28 peer-reviewed papers, 7 (25%) explained the development of the tool, 16 (57%) described evaluation, and 2 (7%) reported both development and evaluation of the tool. The 28 peer-reviewed papers reported on 25 different tools. Of these 25 digital gamified tools, 11 (44%) were web-based tools, 8 (32%) mobile (native mobile or mobile-enabled web) apps, and 6 (24%) virtual reality tools. Overall, tools that were evaluated showed increases in knowledge and intentions to receive vaccines, mixed effects on attitudes, and positive effects on beliefs. We did not observe discernible advantages of one type of digital gamified tool (web based, mobile, virtual reality) over the others. However, a few studies were randomized controlled trials, and publication bias may have led to such positive effects having a higher likelihood of appearing in the peer-reviewed literature. Digital gamified tools appear to have potential for improving vaccine uptake by fostering positive beliefs and increasing vaccine-related knowledge and intentions. Encouraging comparative studies of different features or different types of digital gamified tools could advance the field by identifying features or types of tools that yield more positive effects across populations and contexts. Further work in this area should seek to inform the implementation of gamification for vaccine acceptance and promote effective health communication, thus yielding meaningful health and social impacts.

Automating The Deployment of MERN Stack on AWS App Runner Using AWS Code Pipeline

This paper explores the deployment of web applications utilizing AWS App Runner, a managed platform that simplifies the deployment process by allowing developers to select runtimes and deploy applications without the need for extensive configuration. It focuses on the advantages of using Docker runtime for API execution and web application deployment, highlighting the ease of dockerizing applications and the seamless integration with Amazon Elastic Container Registry (ECR). ECR provides a fully-managed Docker container registry, facilitating the efficient storage, management, and deployment of Docker container images. Additionally, the paper discusses the role of AWS CodeCommit as a secure, scalable, managed source control service for hosting private Git repositories, emphasizing its significance in the deployment pipeline. The combined use of these AWS services offers a streamlined, costeffective solution for deploying scalable and secure web applications directly from source code or container images to the AWS Cloud, thereby enhancing the deployment process for developers.

Mashup-oriented API recommendation via pre-trained heterogeneous information networks

Combining different Web APIs to create Mashups has become very popular nowadays. Choosing suitable ones from massive Web APIs is of vital importance for efficient Mashup creations. A number of Mashup-oriented API recommendation methods have been proposed to address this issue, but they have limitations in their ability to exploit the rich attributes and connection data of Web APIs, which impedes their performance. By modeling the API-related data as a heterogeneous information network and using pre-training technology, this paper proposes an accurate API recommendation method, named PHRec. In this method, the meta paths of APIs in the heterogeneous information network are exploited to obtain their context semantics; the method adopts an attention mechanism. Extensive experiments have been conducted with a real Web API dataset to evaluate the proposed method. The experimental results demonstrate that it significantly outperforms the state-of-the-art methods in the Web API recommendation task.