Compared with normal weather, the traffic flow characteristics changed under condition of adverse which accounted for difference driving behavior. The influence freeway caused by different rainfall intensity was quantified through collecting Hainan Province Eastern Freeway and data. speed density distribution microscopic parameters about rains researched, highway two-regime model weather conditions calibrated. Besides, coefficient that visibility affect determined regression analysis....

Sulfate radical-based advanced oxidation processes exhibit great potential for the degradation of organic pollutants. In this study, P-doped biochar (PBC500) was successfully synthesized by pyrolysis H3PO4-impregnated cow manure waste and employed to activate peroxymonosulfate (PMS) elimination 17β-estradiol (E2). The characterization results showed that surface area, defective structure, functional groups (C=O phosphorus-containing groups) increased after H3PO4 modification. PBC500...

Recently, many Wikipedia-based methods have been proposed to improve the performance of different natural language processing (NLP) tasks, such as semantic relatedness computation, text classification and information retrieval. Among these methods, salient analysis (SSA) has proven be an effective way generate conceptual representation for words or documents. However, its feasibility effectiveness in retrieval is mostly unknown. In this paper, we study how efficiently use SSA performance,...

For software maintenance and evolution, a common problem is to understand how each requirement implemented in the source code. The basic solution of this find fragment code that corresponding implementation requirement. This can be viewed as requirement-slicing - slicing according individual We present an approach set functions main idea our method combine information retrieval technology with static analysis structures. First, we retrieve initial function sets through some model using...

Our investigation into the Affective Reasoning in Conversation (ARC) task highlights challenge of causal discrimination. Almost all existing models, including large language models (LLMs), excel at capturing semantic correlations within utterance embeddings but fall short determining specific relationships. To overcome this limitation, we propose incorporation i.i.d. noise terms conversation process, thereby constructing a structural model (SCM). It explores how distinct relationships fitted...

The sustainable development of docked bikesharing systems has gained focus again owing to several problems in dockless systems, including wanton destruction, theft, illegal parking, loss, and bankruptcy. prediction pickup/return demands is a critical issue for the operation systems. We propose novel local spectral graph convolution (LSGC)- long short-term memory (LSTM) predict based on multi-source data. apply LSGC indicate spatial dependency according geographic information system data that...

Most of the existing information retrieval models assume that terms a text document are independent each other. These integrate three major variables to determine degree importance term for document: within frequency, length and specificity in collection. Intuitively, is not only dependent on aspects mentioned above, but also semantic coherence between document. In this paper, we propose heuristic approach, which query with adopted improve performance. Experimental results standard TREC...

Query expansion is a technique utilized within information retrieval to solve word mismatch between queries and document. In previous method, words are usually selected by counting co-occurrences in the documents. However, not always good indicator for relevance, whereas some background of whole collection. order select words, explicit semantic analysis (ESA) adopted our model estimate two kinds relevance weight. One weight query its relevant extracted from top-ranked documents initial...

The problem of how to partition a software system and thus capture its overall architecture constituent components has become research focus in the community engineering. In literature, many methods have been proposed for solving this problem. For example, both top-down bottom-up based on analyzing graph representation systems proposed. We report an experimental study method method. our study, we capability component capture, recovery time complexity two methods. According results real world...

<abstract><p>Aiming at the problem of solving nonlinear ordinary differential equations with variable coefficients, this paper introduces elastic transformation method into process for first time. A class first-order and a third-order coefficients can be transformed Laguerre equation through transformation. With help general solution equation, these two classes obtained, then curves drawn. This not only expands solvable equations, but also provides new idea...

Bioremediation has drawn widespread concern in passivating heavy metals, but the intense toxicity of metals to biological cells limits application functional strains. Herein, goethite-modified biochar (GMB) was chosen as carrier immobilize phosphate-solubilizing bacteria (PSB) strain L1 for lead and cadmium remediation. Batch experiments showed that GMB-L1 possessed excellent adsorption performance with a maximum 496.54 178.18 mg/g Pb Cd, respectively. Moreover, adding contaminated soil...

In this paper we propose a novel method to estimate the relevance between query and candidate expansion terms for Chinese information retrieval. previous method, are usually selected by counting term co-occurrences in documents. However, not always good indicator relevance, whereas some background of whole collection. order remove noise, an EM-algorithm is used our model two kinds weight. One weight its relevant extracted from top-ranked documents initial retrieval results. The other each...

In this paper we present a Chinese query expansion model based on topic-relevant terms which were acquired from the Google search engine automatically. contrast to earlier methods, our queries are expanded by adding those that most relevant concept of query, rather than selecting terms. Firstly, use automatically extracted short document sets build indexes and in both documents do initial retrieval. Next, acquire Internet top 30 retrieval documents. Finally, expansion. The experiments show...

Open source softwares play an important role in information retrieval research. Most of the existing open systems are implemented Java or C++ programming language. In this paper, we propose Parrot1, a Python-based interactive platform for The proposed has mainly three advantages comparison with systems: (1) It is integrated Jupyter Notebook, which proved to be effective data scientists tackle big and AI problems. As result, users can interactively visualize diagnose model; (2) application...

Incorporating semantic information into document representation is effective and potentially significant to improve retrieval performance. Recently, log-bilinear language model (LBL), as a form of neural model, has been proved be an way learn word representations, but its feasibility effectiveness in mostly unknown. In this paper, we study how efficiently use LBL as-hoc retrieval. We propose (LB-DM) within the modeling framework. The key idea semantically oriented representations for words,...