Modeling Nd<Superscript>3+</Superscript>-Yb<Superscript>3+</Superscript>-Tm<Superscript>3+</Superscript>-Er<Superscript>3+</Superscript> Codoped Telluride Glass Fiber for 0.4 to 2.0 μm Emission Spectra

The modeling of rare-earth-doped fiber amplifier is accomplished by utilizing the rate and propagation equations of distinct levels for a laser medium. A complex theoretical model for neodymium (Nd³⁺), erbium (Er³⁺), thulium (Tm³⁺) and ytterbium (Yb³⁺) codoped telluride glass fiber covering 0.4—2.0 μm emission spectra is presented. The emission spectra of Nd³⁺-Er³⁺-Tm³⁺-Yb³⁺ codoped telluride fiber are realized with the excitation of both 808 and 980 nm lasers pumped at 500mW. Numerical methods are used to calculate the emission spectra covering 0.4—2.0 μm. With the Nd³⁺, Tm³⁺ and Yb³⁺ ion concentrations fixed at 2 × 10²⁰ ion/m³, the Er³⁺ ion concentration optimized to 8 × 1020 ion/m3 and the fiber length spanning from 0.5 to 2 m, a peak amplified spontaneous emission (ASE) power of 19.8mW is attainable, and a minimum ASE power of 7.96mW can also be achieved. The analytical techniques and results indicate that when a telluride codoped fiber with suitable ion concentrations of Nd³⁺, Er³⁺, Tm³⁺ and Yb³⁺ is excited by both 980 and 808 nm pump lasers, 0.4—2.0 μm emission spectra are attainable for vast optical applications.     

Research of Clinical Named Entity Recognition Based on Bi-LSTM-CRF

Electronic Medical Records (EMR) with unstructured sentences and various conceptual expressions provide rich information for medical information extraction. However, common Named Entity Recognition (NER) in Natural Language Processing (NLP) are not well suitable for clinical NER in EMR. This study aims at applying neural networks to clinical concept extractions. We integrate Bidirectional Long Short-Term Memory Networks (Bi-LSTM) with a Conditional Random Fields (CRF) layer to detect three types of clinical named entities. Word representations fed into the neural networks are concatenated by character-based word embeddings and Continuous Bag of Words (CBOW) embeddings trained both on domain and non-domain corpus. We test our NER system on i2b2/VA open datasets and compare the performance with six related works, achieving the best result of NER with F1 value 0.853 7. We also point out a few specific problems in clinical concept extractions which will give some hints to deeper studies.     

Multilingual financial news retrieval and smart recommendation based on big data

In view of the study of finance and economics information, we research on the real-time financial news posted on the authority sites in the world’s major advanced economies. Analyzing the massive financial news of different information sources and language origins, we come up with a basic theory model and its algorithm on financial news, which is capable of intelligent collection, quick access, deduplication, correction and integration with financial news’ backgrounds. Furthermore, we can find out connections between financial news and readers’ interest. So we can achieve a real-time and on-demand financial news feed, as well as provide a theoretical basis and verification of the scientific problems on real-time processing of massive information. Finally, the simulation experiment shows that the multilingual financial news matching technology can give more help to distinguish the similar financial news in different languages than the traditional method.     

Price fluctuation,risk hedge and choice of optimal point

On the question of optimal hedge ratio, this paper firstly draws the chance to choose a market-entering point to the model. Using the replication principle of finance engineering, we make an assumed equity and get the optimal hedge ratio of the model, which gives the theoretical support to the practice. We should not only concern on the market-entering point, but also concern on the period of the hedge that still influences the effect of hedging. But only the period of hedging gives its affection if the time is relatively long.     

Informed trading,heterogeneity investment,liquidity shocks and the valuation of credit default swaps

This paper explores the effect of informed trading, heterogeneity investment and liquidity shocks on the valuation of credit default swaps (CDSs). Under the condition of asymmetric information, the informed trading plays an important role in the valuation of CDS. Instruction order flow has a significant influence on CDS price. And the scope of influence changes in accordance with different time interval, company status and the size of bid-ask spread. Heterogeneity of investors seriously affects the market liquidity and subsequently affects the CDS price. The bigger heterogeneity of the investment philosophy, investment habits, investment preference and so on is the bigger risk for market liquidity, and the higher price for CDS shall be. On the contrary, the conclusion is also consistent. The effectiveness of liquidity, whether it is before or after the financial crisis, dominates the fluctuation of CDS price. The premium of liquidity accounts for 36% to 50% of the CDS price.     

Dynamic response analysis of a multi-column tension-leg-type floating wind turbine under combined wind and wave loading

Floating wind turbines (FWTs) are subjected to combined aerodynamic and hydrodynamic loads varying both in time and amplitude. In this study, a multi-column tension-leg-type FWT (i.e., WindStar TLP system) is investigated for its global performance under normal operating conditions and when parked. The selected variables are analysed using a fully coupled aero-hydro-servo-elastic time domain simulation tool FAST. Three different loading scenarios (wind only, wave only and both combined) are examined to identify the dominant load influencing each response. The key response variables are obtained and compared with those for an NREL 5MW baseline wind turbine installed on land. The results should aid the detailed design of the WindStar TLP system.     

A stochastic concession model for infrastructure projects under build-operate-transfer schemes

Concession period is a key factor in the arrangement of a build-operate-transfer (BOT) contract. This paper proposes a stochastic concession period model for infrastructure projects under a BOT scheme, which takes into account the impact of risks and the risk attitudes of the private investor and the government. A hypothetical case is simulated by using the Monte Carlo simulation method to illustrate the proposed model. The results indicate that even for a project with the same estimations, the concession interval can be different and depends a lot on the risk attitudes of the private investor and the government. The proposed model provides a more reasonable concession interval based on which a specific concession period can be obtained through negotiation between the two parties.     

Urban end distribution optimization under e-commerce environment

With the rapid development of e-commerce, urban end distribution plays more and more important role in e-commerce logistics. The collection and delivery points (CDPs), between online retailers and customers, provide a way to improve the service quality of urban end distribution. But it will be more difficult to obtain an optimal solution of urban end delivery plan when many CDPs joint a complicated delivery network, since the solution space is always too large for many traditional heuristic algorithms to search. In this paper, a two-stage optimization method based on geographic information system (GIS) and improved cooperative particle swarm optimization (CPSO) is proposed. This method takes full advantage of powerful network analysis of GIS and strong global search of CPSO. A new cooperative learning mechanism, global sub-swarm, local sub-swarm and normal sub-swarm (GS-LS-NS), is used to improve the search mode of CPSO. Finally, several experiments are conducted to show the better performance of GIS-CPSO, compared with single PSO, GIS-CPSO and ArcGIS (software of GIS) separately. The conclusion of this research is much useful and applicable for logistics service providers.     

Construction of rate-compatible (RC) low-density parity-check (LDPC) convolutional codes based on RC-LDPC block codes

In this paper, a family of rate-compatible (RC) low-density parity-check (LDPC) convolutional codes can be obtained from RC-LDPC block codes by graph extension method. The resulted RC-LDPC convolutional codes, which are derived by permuting the matrices of the corresponding RC-LDPC block codes, are systematic and have maximum encoding memory. Simulation results show that the proposed RC-LDPC convolutional codes with belief propagation (BP) decoding collectively offer a steady improvement on performance compared with the block counterparts over the binary-input additive white Gaussian noise channels (BI-AWGNCs).     

Evolutionary game on oil and gas companies’ pollution treatment

This paper presents a dynamic model of pollution treatment strategies, driven by the bargaining relationships between governments, oil and gas companies and local residents. The results show that government plays a dominating role and government environmental policies have significant impact on other two parties’ behavior. To achieve better pollution treatment in the future, we put forward the following suggestions. An improvement in the legal system is required, and the legal enforcement needs to be strengthened. In terms of pollution treatment, responsibilities and obligations of these three parties need to be clearly defined. A wellfunctioned reward and penalties system needs to be established for minimizing the environment damage in the future.