Massive Speech Recognition Resource Scheduling System based on Grid Computing

Nowadays, there are too many large-scale speech recognition resources, which makes it difficult to ensure the scheduling speed and accuracy. In order to improve the effect of large-scale speech recognition resource scheduling, a large-scale speech recognition resource scheduling system based on grid computing is designed in this paper. In the hardware part, microprocessor, Ethernet control chip, controller and acquisition card are designed. In the software part of the system, it mainly carries out the retrieval and exchange of information resources, so as to realize the information scheduling of the same type of large-scale speech recognition resources. The experimental results show that the information scheduling time of the designed system is short, up to 2.4min, and the scheduling accuracy is high, up to 90%, in order to provide some help to effectively improve the speed and accuracy of information scheduling.

An Improved Genetic Algorithm on Hybrid Information Scheduling

Background: Patents suggest that efficient hybrid information scheduling algorithm is critical to achieve high performance for heterogeneous multi-core processors. Because the commonly used list scheduling algorithm obtains the approximate optimal solution, and the genetic algorithm is easy to converge to the local optimal solution and the convergence rate is slow. Methods: To solve the above two problems, the thesis proposes a hybrid algorithm integrating list scheduling and genetic algorithm. Firstly, in the task priority calculation phase of the list scheduling algorithm, the total cost of the current task node to the exit node and the differences of its execution cost on different processor cores are taken into account when constructing the task scheduling list, then the task insertion method is used in the task allocation phase, thus obtaining a better scheduling sequence. Secondly, the pre-acquired scheduling sequence is added to the initial population of the genetic algorithm, and then a dynamic selection strategy based on fitness value is adopted in the phase of evolution. Finally, the cross and mutation probability in the genetic algorithm is improved to avoid premature phenomenon. Results: With a series of simulation experiments, the proposed algorithm is proved to have a faster convergence rate and a higher optimal solution quality. Conclusion: The experimental results show that the ICLGA has the highest quality of the optimal solution than CPOP and GA, and the convergence rate of ICLGA is faster than that of GA.

Sistem Penanganan Keluhan Pada Moda Transportasi Kereta Api Studi Kasus: PT. Kereta Api Indonesia Dan Statens Järnvägar Ab, Sweden

Kereta api sebagai moda angkutan massal memiliki karakteristik yang unik. Angkutan ini memiliki kapasitas besar, tingkat keamanan yang tinggi, dan bebas dari kemacetan lalu lintas. Karakteristik tersebut membuat kereta api sebagai angkutan umum utama. Meskipun transportasi kereta api memiliki banyak manfaat bagi kehidupan masyarakat tetapi mereka masih dihadapi oleh masalah. Saat ini operasional kereta api masih diwarnai dengan keterlambatan, kondisi kereta yang tidak baik, dan informasi perjalanan kereta api tidak jelas yang sering merugikan penumpang, dan banyak layanan yang ditawarkan gagal untuk menarik penumpang. Kondisi ini mengakibatkan penurunan kualitas pelayanan dan pengoperasian kerta api tidak memadai. Tujuan dari penelitian ini adalah untuk menganalisis hubungan antara kepuasan konsumen terhadap layanan yang disediakan dengan keinginan untuk melakukan keluhan dan untuk mencari faktor dari kualitas layanan yang memiliki pengaruh yang signifikan antara kepuasan pelanggan terhadap pelayanan PT KAI. Dari data, dan juga studi perbandingan antara PT Kereta Api Indonesia dan Statens Järnvägar AB, Swedia, dapat direkomendasikan desain sistem penanganan pengaduan yang perlu disesuaikan dengan kepentingan konsumen. Penelitian ini menghasilkan beberapa temuan. Pertama, ada enam faktor atribut kualitas layanan yang memiliki pengaruh yang signifikan terhadap kepuasan pelanggan terhadap pelayanan PT KAI untuk kelas komuter (Information, Appearances, Service coverage, Tangible, Safety & security, and Cost). Untuk kelas bisnis memiliki tujuh faktor, (Travel time, Information, Scheduling, Comfort, Tangible, Safety & security, and Service coverage). Dan untuk kelas eksekutif, juga memiliki tujuh faktor yang mempunyai pengaruh signifikan terhadap kepuasan pelanggan, (Appearances, Safety & security, Information, Comfort, Tangible, Travel time, and Cost). Untuk hasil penelitian kedua menunjukkan bahwa untuk kelas komuter atribut Safety & security dan untuk kelas bisnis atribut Information adalah atribut yang memiliki pengaruh signifikan pada keinginan untuk melakukan keluhan. Sementara untuk kelas eksekutif, sebagian besar penumpang puas dengan pelayanan yang diberikan oleh PT. KAI. Untuk pertanyaan penelitian ketiga, untuk mengurangi jumlah keluhan, maka harus mengambil beberapa mekanisme yang efektif untuk menangani pengaduan tersebut belajar dari sistem yang diterapkan oleh SJ. AB.Kata kunci: Pelayanan keluhan penumpang, PT Kereta Api Indonesia, Statens Järnvägar ABRead phonetically

Big Data Analysis and Simulation of Distributed Marine Green Energy Resources Grid-Connected System

In order to improve the working stability of distributed marine green energy resources grid-connected system, we need the big data information mining and fusion processing of grid-connected system and the information integration and recognition of distributed marine green energy grid-connected system based on big data analysis method, and improve the output performance of energy grid-connected system. This paper proposed a big data analysis method of distributed marine green energy resources grid-connected system based on closed-loop information fusion and auto correlation characteristic information mining. This method realized the big data closed-loop operation and maintenance management of grid-connected system, and built the big data information collection model of marine green energy resources grid-connected system, and reconstructs the feature space of the collected big data, and constructed the characteristic equation of fuzzy data closed-loop operation and maintenance management in convex spaces, and used the adaptive feature fusion method to achieve the auto correlation characteristics mining of big data operation and maintenance information, and improved the ability of information scheduling and information mining of distributed marine green energy resources grid-connected system. Simulation results show that using this method for the big data analysis of distributed marine green energy resources grid-connected system and using the multidimensional analysis technology of big data can improve the ability of information scheduling and information mining of distributed marine green energy resources grid-connected system, realizing the information optimization scheduling of grid-connected system. The output performance of grid connected system has been improved.