Traditional news recommendation methods usually encode news text and user interests directly into vectors and then calculate the matching degree of the two. Such approaches may lose low-level matching signals and diff...
Traditional news recommendation methods usually encode news text and user interests directly into vectors and then calculate the matching degree of the two. Such approaches may lose low-level matching signals and different levels of textual information. We propose a BERT-based news recommendation model, MLFBERT. This model consists of a BERT-based news encoder and a user encoder. The news encoder uses a PLM Fusion Module and additive attention to generate embeddings of available news items. The user encoder builds a user profile by aggregating the representations of their news browsing history. Recommendations are made by computing matching scores between the user profile and candidate news. Experiments on the MIND news dataset demonstrate that our model outperforms existing methods on multiple evaluation metrics.
This paper presents details of studies conducted to investigate interpretable and explainable machine learning and AI models for cardiovascular disease detection based on the publicly available Cleveland dataset. The ...
详细信息
This paper presents details of studies conducted to investigate interpretable and explainable machine learning and AI models for cardiovascular disease detection based on the publicly available Cleveland dataset. The study involves evaluating the interpretability and explainability capabilities of tradition shallow machine learning models and their potential for implementation under low resource settings, with limited training data available for model building, as compared to high performing deep learning models, requiring massive training datasets.
In recent years, computer architecture instruction has undergone significant transformation, moving away from a primarily theoretical approach, and placing greater emphasis on practical, hands-on learning. Leading thi...
In recent years, computer architecture instruction has undergone significant transformation, moving away from a primarily theoretical approach, and placing greater emphasis on practical, hands-on learning. Leading this shift is the Raspberry Pi (RaPi), a credit card-sized computer powered by the versatile Broadcom BCM2835 system-on-a-chip (SoC). RaPi offers an open-source operating system, an extensive software library, and abundant online documentation, creating an ideal environment for dynamic and experiential learning. This paper features the successful integration of RaPi into the computer architecture curriculum at CNU, with a primary focus on hands-on labs. These labs cover diverse topics, including RaPi setup and connectivity, low-level programming, Python-based sensor interfacing, and AI-driven image processing for object detection and recognition. The paper provides valuable insights for educators looking to incorporate this single-board computer into their teaching methods, highlighting its multifaceted applications in AI and computer architecture to enhance students' learning experience.
The determination of engineering project cost (PC) is one of the important contents of engineering project management, but the estimation of engineering cost (EC) generally has the disadvantages of large error and lon...
详细信息
Students should engage in creative works in schools. However, there is no sensor to measure on the spot whether they engage in creative works. This work assumes engagement in creative works shows up in their writing f...
详细信息
Students should engage in creative works in schools. However, there is no sensor to measure on the spot whether they engage in creative works. This work assumes engagement in creative works shows up in their writing features with pens. It uses a tablet and an electronic pen to acquire the writing features. It applies Random forests to detect the writing features. The results have showed that the features appear in writing pressure, velocity, and pen tilt angle.
With the continuous development of network technology as well as science and technology, artificial intelligence technology and its related scientific and technological applications, in this process, were born. Among ...
详细信息
In this information age, more and more data are being produced every day from different sectors. Healthcare is no exception where huge amounts of data is being produced daily. However, this data is not effective. As t...
In this information age, more and more data are being produced every day from different sectors. Healthcare is no exception where huge amounts of data is being produced daily. However, this data is not effective. As the phrase goes “data is the new oil”, it is important to acknowledge the importance of data and how it can make work easier for professionals across fields. For the healthcare sector, the various data that are collected can be used to extract knowledge to make diagnosis of certain diseases easier. One such disease which can be predicted is the heart disease which is becoming a top killer amongst the populations across the world. This paper will investigate how machine learning (ML) algorithms can be used to predict heart diseases using heart disease data and will compare the 3 machine learning algorithms - decision tree, naive bayes, and random forest to determine the most accurate model for predicting this deadly condition.
The ubiquitous use of Internet of Things devices in supply chain management has facilitated manufacturers and other stakeholders to monitor various aspects of the product or shipment. However, conventional solutions h...
The ubiquitous use of Internet of Things devices in supply chain management has facilitated manufacturers and other stakeholders to monitor various aspects of the product or shipment. However, conventional solutions have numerous challenges ranging from data integrity, product provenance, and trust issues to scalability problems. This research demonstrates how scalability and provenance of cement bags can be ensured through a combination of blockchain technology and the Internet of Things devices. The distributed ledger securely records each step, from acquiring raw materials to delivering the finished product. The inherent transparency of blockchain technology promotes confidence among all the stakeholders in the supply chain and makes it easier to comply with legal requirements. The security and performance analysis determines that the proposed solution is secure and efficient with minimal performance overheads.
A prototype project proof-of-concept involving pest control and eradication using autonomous robots in scaled down simulated arena. This simulation allows the design and testing of autonomous hunter robots that can se...
详细信息
A prototype project proof-of-concept involving pest control and eradication using autonomous robots in scaled down simulated arena. This simulation allows the design and testing of autonomous hunter robots that can seek out, control and eradicate known pest. This paper describes the test arena behind the scaled down simulation.
There are many existing applications for solving CPU scheduling problems. However, these applications are suffering from some defects such as they didn't cover all the scheduling algorithms, didn't cover preem...
There are many existing applications for solving CPU scheduling problems. However, these applications are suffering from some defects such as they didn't cover all the scheduling algorithms, didn't cover preemptive and non-preemptive scheduling, and most importantly, they are missing some calculations. In this work, we proposed CPU Scheduling Solver (CSS), an android application that targets the students to help them solve any CPU scheduling problem using the existing scheduling algorithms: First Come First Served (FCFS), non-preemptive Shortest Job First (SJF), preemptive SJF, and Round Robin (RR). The evaluation of the CSS application performance proves that the proposed application works effectively by computing the evaluation metrics for any solving algorithm; then draw its Gantt chart correctly. The CSS application is available in https://***/store/apps/details?id=***&pcampaignid=web_share
暂无评论