Search

In this work, we study an important task in location-based services, namely Personalized Route Recommendation (PRR). Given a road network, the PRR task aims to generate user-specific route suggestions for replying to users' route queries. A classic approach is to adapting search algorithms to construct pathfinding-like solutions. These methods typically focus on reducing search space with suitable heuristic strategies. For these search algorithms, heuristic strategies are often handcrafted, which are not flexible to work in complicated task settings. In this paper, we propose to improve search algorithms with neural networks for solving the PRR task based on the widely used A* algorithm. Our model consists of two components. First, we employ attention-based Recurrent Neural Networks (RNN) to model the cost from the source to the candidate location by incorporating useful context information. Second, we propose to use an estimation network for predicting the cost from a candidate location to the destination. The two components are integrated in a principled way for deriving a more accurate cost of a candidate location for the A* algorithm. Extensive experiment results on three real-world datasets have shown the effectiveness and robustness of the proposed model.

Aim Sedentary behaviors are activities that expend less energy than 1.5 metabolic equivalents, such as watching TV, sitting. We aim to systematically review the evidence to determine the association of sedentary behavior (SB) with the risk of gestational diabetes mellitus, and the associations of SB with health outcomes in pregnant women with gestational diabetes mellitus (GDM) and women at high risk of gestational diabetes mellitus. Methods Two researchers independently performed a literature search using PubMed, Medline, EMBASE, CINAHL, the Cochrane Library, the Rehabilitation and Sports Medicine Source, and the Sedentary Behavior Research Database for studies published up to January 27, 2021. Randomized controlled trials, quasi-experimental studies, and observational studies were screened and selected if SB was assessed in the context of gestational diabetes mellitus. Two reviewers extracted the relevant information and independently assessed the included studies for quality and bias. Results Eighteen studies were included, involving 60 804 patients, and the timeline of the original study included was from 2006 to 2021. Our review showed that SB before pregnancy was not associated with the risk of gestational diabetes mellitus, while SB during pregnancy was uncertain about the risk of gestational diabetes mellitus. In addition, there were significant associations between SB time and metabolic outcomes. In addition, SB may cause some adverse maternal and fetal outcomes. Conclusion SB is likely to be the intermediary variable between occurrence of GDM and adverse maternal and infant outcome, rather than a direct influencing factor. Limited by the quantity and quality of included studies, the above conclusions need to be verified by more high-quality studies.

In recent years, customized bus (CB), as a complementary form of urban public transport, can reduce residents’ travel costs, alleviate urban traffic congestion, reduce vehicle exhaust emissions, and contribute to the sustainable development of society. At present, customized bus travel demand information collection method is passive. There exist disadvantages such as the amount of information obtained is less, the access method is relatively single, and more potential travel demands cannot be met. This study aims to combine mobile phone signaling data, point of interest (POI) data, and secondary property price data to propose a method for identifying the service areas of commuter CB and travel demand. Firstly, mobile phone signaling data is preprocessed to identify the commuter’s location of employment and residence. Based on this, the time-space potential model for commuter CB is proposed. Secondly, objective factors affecting commuters’ choice to take commuter CB are used as model input variables. Logistic regression models are applied to estimate the probability of the grids being used as commuter CB service areas and the probability of the existence of potential travel demand in the grids and, further, to dig into the time-space distribution characteristics of people with potential demand for CB travel and analyze the distribution of high hotspot service areas. Finally, the analysis is carried out with practical cases and three lines are used as examples. The results show that the operating companies are profitable without government subsidies, which confirms the effectiveness of the method proposed in this paper in practical applications.

Objective To study the application of a machine learning algorithm for predicting gestational diabetes mellitus (GDM) in early pregnancy. Methods This study identified indicators related to GDM through a literature review and expert discussion. Pregnant women who had attended medical institutions for an antenatal examination from November 2017 to August 2018 were selected for analysis, and the collected indicators were retrospectively analyzed. Based on Python, the indicators were classified and modeled using a random forest regression algorithm, and the performance of the prediction model was analyzed. Results We obtained 4806 analyzable data from 1625 pregnant women. Among these, 3265 samples with all 67 indicators were used to establish data set F1; 4806 samples with 38 identical indicators were used to establish data set F2. Each of F1 and F2 was used for training the random forest algorithm. The overall predictive accuracy of the F1 model was 93.10%, area under the receiver operating characteristic curve (AUC) was 0.66, and the predictive accuracy of GDM-positive cases was 37.10%. The corresponding values for the F2 model were 88.70%, 0.87, and 79.44%. The results thus showed that the F2 prediction model performed better than the F1 model. To explore the impact of sacrificial indicators on GDM prediction, the F3 data set was established using 3265 samples (F1) with 38 indicators (F2). After training, the overall predictive accuracy of the F3 model was 91.60%, AUC was 0.58, and the predictive accuracy of positive cases was 15.85%. Conclusions In this study, a model for predicting GDM with several input variables (e.g., physical examination, past history, personal history, family history, and laboratory indicators) was established using a random forest regression algorithm. The trained prediction model exhibited a good performance and is valuable as a reference for predicting GDM in women at an early stage of pregnancy. In addition, there are certain requirements for the proportions of negative and positive cases in sample data sets when the random forest algorithm is applied to the early prediction of GDM.

The fuzzy cognitive map (FCM) is a powerful model for system state prediction and interpretable knowledge representation. Recent years have witnessed the tremendous efforts devoted to enhancing the basic FCM, such as introducing temporal factors, uncertainty or fuzzy rules to improve interpretation, and introducing fuzzy neural networks or wavelets to improve time series prediction. But how to achieve high-precision yet interpretable prediction in cross-domain real-life applications remains a great challenge. In this article, we propose a novel FCM extension called deep FCM (DFCM) for multivariate time series forecasting, in order to take both the advantage of FCM in interpretation and the advantage of deep neural networks in prediction. Specifically, to improve the predictive power, DFCM leverages a fully connected neural network to model connections (relationships) among concepts in a system, and a recurrent neural network to model unknown exogenous factors that have influences on system dynamics. Moreover, to foster model interpretability encumbered by the embedded deep structures, a partial derivative-based approach is proposed to measure the connection strengths between concepts in DFCM. An alternate function gradient descent algorithm is then proposed for parameter inference. The effectiveness of DFCM is validated over four publicly available datasets with the presence of seven baselines. DFCM indeed provides an important clue to building interpretable predictors for real-life applications.

Cancer stem cells (CSCs) are primarily maintained by a network of pluripotency transcription factors (PTFs). Given a close relationship of CSC regulation with epidermal growth factor receptor and vascular endothelial growth factor signalling, we investigated whether single-nucleotide polymorphisms (SNPs) in PTF genes are related to the efficacy of cetuximab and/or bevacizumab in patients with metastatic colorectal cancer (mCRC).Genomic and clinical data from three independent clinical trial cohorts were tested: cetuximab cohort (FOLFIRI/cetuximab arm in FIRE-3, n = 129), bevacizumab cohort 1 (FOLFIRI/bevacizumab arm in FIRE-3, n = 107) and bevacizumab cohort 2 (FOLFIRI/bevacizumab arm in TRIBE, n = 215). Genomic DNA extracted from blood samples was genotyped, and ten SNPs were tested for association with clinical outcomes.In the cetuximab cohort, four SNPs were significantly associated with progression-free survival in univariate analysis: NANOG rs11055767 (any A allele vs C/C, hazard ratio [HR] = 0.62, 95% confidence interval [CI] = 0.42-0.94, p = 0.02), NANOG rs10744044 (any A allele vs G/G, HR = 0.59, 95% CI = 0.39-0.90, p = 0.01), NANOGP8 rs2168958 (any C allele vs A/A, HR = 2.12, 95% CI = 1.36-3.29, p 0.001) and NANOGP8 rs2279066 (any C allele vs T/T, HR = 1.80, 95% CI = 1.06-1.68, p = 0.03). Multivariate analysis confirmed the significant associations for NANOGP8 rs2168958 and NANOGP8 rs2279066. In either bevacizumab cohort, no significant associations were observed in univariate analysis.Germ line polymorphisms in the PTF genes could be predictive markers for cetuximab in mCRC.

Recent years have witnessed the emergence of worldwide megalopolises and the accompanying public safety events, making urban safety a top priority in modern urban management. Among various threats, dangerous goods such as gas and hazardous chemicals transported through cities have bred repeated tragedies and become the deadly “bomb” we sleep with every day. While tremendous research efforts have been devoted to dealing with dangerous goods transportation (DGT) issues, further study is still in great need to quantify this problem and explore its intrinsic dynamics from a big data perspective. In this article, we present a novel system called DGeye , to feature a fusion between DGT trajectory data and residential population data for dangers perception and prediction. Specifically, DGeye first develops a probabilistic graphical model-based approach to mine spatio-temporally adjacent risk patterns from population-aware risk trajectories. Then, DGeye builds the novel causality network among risk patterns for risk pain-point identification, risk source attribution, and online risky state prediction. Experiments on both Beijing and Tianjin cities demonstrate the effectiveness of DGeye in real-life DGT risk management. As a case in point, our report powered by DGeye successfully drove the government to lay down gas pipelines for the famous Guijie food street in Beijing.

Big data analytics (BDA) is regarded as an advanced tool for achieving sustainable development as part of the grand challenges (GCs). However, it is not clear how BDA can be used by data scientists to solve the GCs with multisource data in a cross-disciplinary approach. Based on a case study of city-based dangerous goods transportation (DGT), this paper explores how data scientists use BDA to triangulate data, methods, knowledge and solutions for solving GCs. The contribution of this study is threefold: (1) it contributes to research on GCs and discusses how BDA can be used in problem solving for multidomain GCs from a management perspective; (2) it enriches the theory of information triangulation and proposes several steps for information triangulation in BDA to solve GCs; and (3) it contributes some practical implications for the management of organizations when solving social problems and pursuing sustainable development.

ABTRACT Background Atopic dermatitis (AD) is a heritable and heterogeneous inflammatory chronic skin disorder. Utilizing decision tree/supervised learning of extensive clinical, molecular and genetic data, we aimed to define distinct AD endotypes. Methods Deep phenotyping and whole-genome sequencing was performed on samples obtained from participants of EPIONE, a randomized-controlled phase III study in AD patients with severe pruritus comprising mild (23%), moderate (64%) and severe (13%) AD as determined by AD Investigator Global Assessment scale. Three categories of analysis were performed: clinical associations, lab value associations (EOS, IgE, cytokines) and genetic analysis of whole-genome sequencing data Results Based on a decision tree, we found that five clinical features from the SCORing Atopic Dermatitis (SCORAD) Index can accurately differentiate between IGA severities. We observe a significant difference between severity and eosinophil counts (p Conclusion Our results suggest significant differences between severity groups across a number of features appear to constitute distinct endotypes with likely distinct causative factors. Differing underlying pathophysiology’s indicate endotype knowledge is critical to help guide therapeutic approaches to AD. Capsule summary AD is a heritable and heterogeneous skin disorder that makes the ‘one size fits all’ therapeutic approach suboptimal for patients with AD. We attempted to define AD endotypes based on clinical, molecular, and genetic characteristics. Clinical, CBC, and genetic associations all tend to suggest existence of separate endotypes.

Background BRCA1/2 mutations have been shown to be associated with the development of many solid tumors including endometrial cancer (EC). The objectives of this study are to analyze the association between BRCA1/2 mutational status and clinicopathological characteristics as well as outcomes in EC patients. Methods 510 eligible EC patients from the Cancer Genome Atlas database were included in the study. The association between clinicopathological characteristics and different BRCA1/2 mutational status was compared and analyzed. Analyses of the impact of BRCA mutations on survival in EC patients was conducted using Kaplan-Meier survival analyses and Cox regressions. In order to control confounding bias between groups, propensity score matching method was used. Results Among the eligible patients, 11 (2.2%) harbored BRCA1 mutation, 43 (8.4%) harbored BRCA2 mutation, and 36 (7.1%) harbored both. Body mass index, rates of hypertension history, proportion of non-endometrioid histology and rates of positive peritoneal cytology were lower in patients with BRCA1/2 mutations compared with the group of wild-type counterpart (p = 0.020, 0.048, 0.001 and 0.012, respectively). Patients with BRCA1/2 mutations showed longer overall (OS) and recurrence-free survival (RFS) (in Kaplan-Meier analyses, p Conclusion BRCA1/2 mutations are associated with better survival in patients with high grade endometrioid EC, indicating the value of BRCA testing in EC clinical management.

Tunnel excavations are commonly carried out using the drill and blast method, which may cause blast-induced damage to adjacent buildings. Peak particle velocity (PPV) is a widely used parameter for evaluating the damage of blasting vibration. However, accurately predicting PPV is difficult with traditional empirical predicting methods because their results are often different from actual conditions. In this study, attenuation formula of propagation velocity of elastic stress waves in elastomer is derived on the basis of stress wave theory. Moreover, the formula for predicting PPV is modified in the case of multihole and multistage blasting and then applied to Guanlinzi Tunnel, which downtraverses through National Highway 316. Results show that the modified formula obtains a small relative error between predicted and in situ monitoring PPVs and can properly reflect the propagation law of PPV under the condition of multihole and multistage blasting. This work has important application prospects and can provide a reference for similar excavation blasting and vibration control methods.

Aim: To identify the psychological change process of the registered nurses who worked in the epicenter of the COVID-19 outbreak.Background: The pandemic of COVID-19 has continued to pose an unprecedented threat and challenge to people's health around the world. Nurses are at high risk because they work within the closest proximity to patients. Understanding nurses' psychological change process during the care for patients with COVID-19 is imperative for healthcare leaders.Methods: This was a qualitative descriptive study that took place in a hospital in Wuhan, China, the epicenter of the COVID-19 epidemic, from February 9th to March 15th, 2020. Using purposive sampling, we interviewed 23 nurses. Data were analyzed using Colaizzi's method of data analysis to find, understand, and describe nurses' experiences.Results: The psychological change process of frontline nurses included three stages, early, middle, and later stages. The psychological characteristics of each period were ambivalence, emotional exhaustion, and energy renewal, respectively. Nurse leaders were anchors in facilitating frontline nurses' psychological adaptation.Conclusions: In the past month, the psychological characteristics of nurses changed over time. The study indicated the necessity for nurse leaders to implement intervention programs based on nurses' psychological characteristics in different periods to promote nurses' health during this critical time period.

The genetic background of Atopic Dermatitis (AD) with chronic pruritus is complex. Filaggrin (FLG) is an essential gene in the epidermal barrier formation s. Loss-of-function (LOF) variants in FLG associated with skin barrier dysfunction constitute the most well-known genetic risk factor for AD. In this study, we focused on the frequency and effect of FLG loss-of-function variants in association with self-reported age-of-onset of AD. The dataset consisted of 386 whole-genome sequencing (WGS) samples. We observe a significant association between FLG LOF status and age-of-onset, with earlier age of onset of AD observed in the FLG LOF carrier group (p-value 0.0003, Wilcoxon two-sample test). We first tested this on the two most prevalent FLG variants. Interestingly, the effect is even stronger when considering all detected FLG LOF variants. Having two or more FLG LOF variants associates with the onset of AD at 2 years of age. In this study, we have shown enrichment of rare variants in the EDC region in cases compared with controls. Age-of-onset analysis shows not only the effect of the FLG and likely EDC variants in terms of the heightened risk of AD, but foremost enables to predict early-onset, lending further credence to the penetrance and causative effect of the identified variants. Understanding the genetic background and risk of early-onset is suggestive of skin barrier dysfunction etiology of AD with chronic pruritus

CyberKnife stereotactic body radiation therapy (SBRT) is becoming increasingly used for cancer treatment and, to maximize its clinical application, it is important to define the dosimetric characteristics, optimal dose and fractionation regimens. The aim of this study was to evaluate the dose fall-off in two fractionated regimens of CyberKnife SBRT during the treatment of thoracic spinal metastasis. Patients with spinal metastasis involving a vertebra and pedicle were treated with 40 Gy in 5 fractions (n = 4), and patients with spinal metastasis involving only a vertebra received 33 Gy in 3 fractions (n = 4). A new approach was used to measure absolute dose fall-off distance, relative dose fall-off distance, and the dose fall-off per unit distance along four reference directions in the axial plane. Patients treated with 33 Gy/3 fractions had a greater absolute dose fall-off distance in direction 1 (from the point with maximum dose [Dmax] towards the spinal cord) and direction 3 (the opposite of the direction 1), a greater relative dose fall-off distance in direction 3, and a lower dose fall-off per unit distance in direction 1 and 3 compared to patients treated with 40 Gy/5 fractions (all p < 0.05). Overall, the dose fall-off towards the spinal cord is rapid during the treatment of thoracic spinal metastasis with CyberKnife SBRT, which allows a higher dose of radiation to be delivered to the tumor and, at the same time, better protection of the spinal cord.

Packaging is one of the most important procedures of the fabrication of photoelectric devices, such as ZnMgO solar blind ultraviolet (UV) photodetectors. The roles of packaging are: electrically interconnecting an unpackaged photodetector and pin as well as fixing it, and protecting the unpackaged photodetector from external dirt, humidity and atmosphere. Nowadays, the research on the ZnMgO solar-blind UV photodetectors has well developed. However, the research on the packaging of these photodetectors are rare. In this study, we have focused on the packaging of ZnMgO solar blind UV photodetectors. After a direct wire bonding method between steel wires and unpackaged photodetectors by conductive silver adhesives and a sealing treatment by silica gel, the ZnMgO photodetector was packaged, and the responsivity of the device was found to increase after the packaging. A −3 dB cutoff wavelength of 295 nm and a responsivity of 191 A W−1 were achieved under a bias of 10 V. The 90–10% decay time of the device was 280 ms, and the dark current of the device was 115 pA under a bias of 10 V. Moreover, the stability of the ZnMgO-based solar blind photodetector obviously improved due to the well-sealed packaged structure.

International audience; The rapidly-growing business of ride-on-demand (RoD) service such as Uber, Lyft and Didi proves the effectiveness of their new service model-using mobile apps and dynamic pricing to coordinate between drivers, passengers and the service provider, to manipulate the supply and demand, and to improve service responsiveness as well as quality. Despite its success, dynamic pricing creates a new problem for drivers: how to seek for passengers to maximize revenue under dynamic prices. Seeking route recommendation has already been studied extensively in traditional taxi service, but most studies do not consider the effects of taxis and passengers on the seeking taxi simultaneously. Further, in RoD service it is necessary to consider more factors such as dynamic prices, the status of other transportation services, etc. In this paper, we employ a force-directed approach to model, by analogy, the relationship between vacant cars and passengers as that between positive and negative charges in electrostatic field. We extract features from multi-source urban data to describe dynamic prices, the status of RoD, taxi and public transportation services, and incorporate them into our model. The model is then used in route recommendation in every intersection so that a driver in a vacant RoD car knows which road segment to take next. We conduct extensive experiments based on our multi-source urban data, including RoD service operational data, taxi GPS trajectory data and public transportation distribution data, and results not only show that our approach outperforms existing baselines, but also justify the need to incorporate multi-source urban data and dynamic prices.

This qualitative study describes the psychological experience of patients hospitalized with COVID-19. These patients went through 3 psychological stages: extremely uncertainties during the initial diagnostic stage, complicated feelings of negativity during the treatment stage, and positive growth in the recovery stage. It is important for nurses to provide holistic care.

With the increase of traffic prediction models, there has become an urgent need to develop a standardized framework to implement and evaluate these methods. This paper presents LibCity, a unified, comprehensive, and extensible library for traffic prediction, which provides researchers with a credible experimental tool and a convenient development framework. In this library, we reproduce 42 traffic prediction models and collect 29 spatial-temporal datasets, which allows researchers to conduct comprehensive experiments in a convenient way. To accelerate the development of new models, we design unified model interfaces based on unified data formats, which effectively encapsulate the details of the implementation. To verify the effectiveness of our implementations, we also report the reproducibility comparison results of LibCity, and set up a performance leaderboard for the four kinds of traffic prediction tasks. Our library will contribute to the standardization and reproducibility in the field of traffic prediction. The open source link of LibCity is https://github.com/LibCity/Bigscity-LibCity.

BACKGROUND: The TIMELESS–TIPIN complex protects the replication fork from replication stress induced by chemotherapeutic drugs. We hypothesised genetic polymorphisms of the TIMELESS–TIPIN complex may affect the response, progression-free survival (PFS), and overall survival (OS) of cytotoxic drugs in patients with metastatic colorectal cancer (mCRC). METHODS: We analysed data from the MAVERICC trial, which compared FOLFOX/bevacizumab and FOLFIRI/bevacizumab in untreated patients with mCRC. Genomic DNA extracted from blood samples was genotyped using an OncoArray. Eight functional single nucleotide polymorphisms (SNPs) in TIMELESS and TIPIN were tested for associations with clinical outcomes. RESULTS: In total, 324 patients were included (FOLFOX/bevacizumab arm, n = 161; FOLFIRI/bevacizumab arm, n = 163). In the FOLFOX/bevacizumab arm, no SNPs displayed confirmed associations with survival outcomes. In the FOLFIRI/bevacizumab arm, TIMELESS rs2291739 was significantly associated with OS in multivariate analysis (G/G vs. any A allele, hazard ratio = 3.06, 95% confidence interval = 1.49–6.25, p = 0.004). TIMELESS rs2291739 displayed significant interactions with treatment regarding both PFS and OS. CONCLUSIONS: TIMELESS rs2291739 might have different effects on therapeutic efficacy between oxaliplatin- and irinotecan-based chemotherapies. Upon further validation, our findings may be useful for personalised approaches in the first-line treatment of mCRC.

We conducted an observational research study to collect information on sleep–wake patterns from participants with a confirmed status of the cryptochrome circadian clock 1 (CRY1) splicing variant, CRY1Δ11 c.1657 + 3A > C, and their controls, defined as wild-type (WT) family members. Altogether, 67 participants were enrolled and completed this study in Turkey, recruited from a list of families with at least one CRY1-confirmed member. We measured sleep–wake patterns and metabolic output, specifically time and frequency of bowel movements, for all participants by daily post-sleep diaries over 28 days. The sleep diary measured self-reported bed time, wake time, midpoint of sleep, and latency to persistent sleep (LPS), and accounted for naps and awakenings for religious purposes. Wake time and midpoint of sleep were significantly later in the CRY1Δ11 variant group versus WT, and LPS was significantly greater in participants in the CRY1Δ11 variant group. The mean bed time on all nights of sleep was later in participants with a CRY1Δ11 variant versus WT. Additionally, participants with a CRY1Δ11 variant had significantly affected metabolic outputs, measured by later bowel movements than WT participants. These results demonstrate that, on average, individuals with the studied splicing variant experience pronounced delays in sleep period and circadian-related metabolic processes.

Ultraviolet (UV) communication has been the topic of extensive recent research due to its non-line-of-sight (NLOS) communication, anti-interference, and high confidentiality. Given the lack of the UV Media Access Control (MAC) protocol, this paper proposes a UV lossless contention MAC (UVLLC-MAC) protocol creatively. This MAC protocol is based on the superposition logic of UV power, which effectively reduces the collision loss of multi-node contention access. The basic working mechanism and protocol flow are given in detail, and the network performance is simulated and analyzed mathematically based on a four-node UV network. Comparing with the slotted ALOHA protocol, the simulation results indicate that the network has a higher throughput and lower delay under the protocol. The practical networking of four communication terminals is designed and implemented, and the effectiveness of the new UVLLC-MAC protocol is fully verified.

BACKGROUND: Metastasis and recurrence are the main causes of death from cervical cancer (CC), thus it is important to identify more effective biomarkers to improve its prognosis. The purpose of our research was to determine the potential role of autophagy-related long non-coding RNA (lncRNA) in CC and to construct an autophagy-related lncRNA signature for survival of CC. METHODS: The lncRNAs in CC were downloaded from The Cancer Genome Atlas (TCGA) database, and autophagy-related lncRNAs were identified through the co-expression of lncRNA genes and autophagy genes. Several autophagy-related lncRNAs with prognostic value (AC012306.2, AL109976.1, ATP2A1-AS1, ILF3-DT, Z83851.2, STARD7-AS1, AC099343.2, AC008771.1, DBH-AS1, and AC097468.3) were identified using univariate and multivariate Cox regression analyses and a prognostic signature was established. The signature effect was detected by univariate Cox regression analysis [hazard ratio (HR) =1.665; 95% confidence interval (CI): 1.331–2.082; P

We build a deep-learning-based SEIR-AIM model integrating the classical Susceptible-Exposed-Infectious-Removed epidemiology model with forecast modules of infection, community mobility, and unemployment. Through linking Google's multi-dimensional mobility index to economic activities, public health status, and mitigation policies, our AI-assisted model captures the populace's endogenous response to economic incentives and health risks. In addition to being an effective predictive tool, our analyses reveal that the long-term effective reproduction number of COVID-19 equilibrates around one before mass vaccination using data from the United States. We identify a "policy frontier" and identify reopening schools and workplaces to be the most effective. We also quantify protestors' employment-value-equivalence of the Black Lives Matter movement and find that its public health impact to be negligible., Preprint, not

Asymptomatic individuals carrying SARS-CoV-2 can transmit the virus and contribute to outbreaks of COVID-19, but it is not yet clear how the proportion of asymptomatic infections varies by age and geographic location. Here we use detailed surveillance data gathered during COVID-19 resurgences in six cities of China at the beginning of 2021 to investigate this question. Data were collected by multiple rounds of city-wide PCR test with detailed contact tracing, where each patient was monitored for symptoms through the whole course of infection. We find that the proportion of asymptomatic infections declines with age (coefficient =-0.006, P60 years. Using an age-stratified compartment model, we show that this age-dependent asymptomatic pattern together with the age distribution of overall cases can explain most of the geographic differences in reported asymptomatic proportions. Combined with demography and contact matrices from other countries worldwide, we estimate that a maximum of 22%–55% of SARS-CoV-2 infections would come from asymptomatic cases in an uncontrolled epidemic based on asymptomatic proportions in China. Our analysis suggests that flare-ups of COVID-19 are likely if only adults are vaccinated and that surveillance and possibly control measures among children will be still needed in the future to contain epidemic resurgence.

Simple Summary Circulating tumor cell (CTC) detection is a non-invasive and easy way to collect and separate single cells to use as tumor biomarkers. In this article, we show the feasibility and clinical value of RNA sequencing of CTCs at the single-cell level by exploiting CTC-FIND with SMART-Seq v4 technologies in patients with metastatic colorectal cancer (mCRC). CTCs were classified by epithelial, epithelial-mesenchymal transition (EMT), and stem cell-related gene expression. A patient who had EMT gene expressing CTCs showed significantly shorter PFS and OS regardless of epithelial CTCs presence. Therefore, the presence of CTCs expressing EMT-related genes at the single-cell level may be ofprognostic value and a potential biomarker for risk stratification in patients with refractory mCRC who are considered to have highly heterogeneous features. Abstract Background: Circulating tumor cells (CTCs) are a prognostic marker in patients with metastatic colorectal cancer (mCRC). However, little is known about the characterization of CTCs in mCRC at the single-cell level using RNA sequencing. The purpose of this study was to validate the capability to detect and isolate single CTCs for single-cell RNA sequencing (scRNA-seq) and to identify clinical significance at a single CTC level. Methods: Single CTCs from 27 mCRC patients were collected by CTC-FIND, which is comprised of filter separation and immunomagnetic depletion to collect ultra-pure CTC samples. To address tumor heterogeneity, CTCs were collected without relying on any traditional CTC markers, such as epithelial and mesenchymal cell antigens, and were undertaken by scRNA-seq using SMART-Seq v4. Results: We identified 59 single CTCs which were classified into four groups by epithelial, epithelial-mesenchymal transition (EMT) and stem cell-related gene expression. Patients receiving second or later-line treatment who had EMT gene expressing CTCs had a significantly shorter PFS and OS. Conclusions: Exploiting CTC-FIND with SMART-Seq v4 showed that scRNA-seq of CTCs may shed new insight into tumor heterogeneity of mCRC and that the presence of CTCs expressing EMT-related genes at the single-cell level could have prognostic value in mCRC patients.

BackgroundAccurate assessment of predictive biomarker expression is critical in patient selection in clinical trials or clinical practice. However, changes in biomarker expression may occur after treatment. The aim of the present study was to evaluate the effects of chemotherapy on MET expression in gastric cancer (GC).MethodsMET expression was examined immunohistochemically before and after treatment in 122 patients with unresectable or recurrent GC, and was evaluated according to H-score or the scoring criteria used in the MetMAb trial. MET gene amplification was assessed by chromogenic in situ hybridization (CISH). The antitumor effect of MET targeted therapy was investigated in human gastric cancer cells in vitro and in vivo, and the underlying molecular mechanisms were analyzed by western blot.ResultsMET expression was associated with Lauren classification as well as tumor differentiation by either scoring system. MET amplification was not associated with clinical characteristics. Of the 71 patients who had paired pre- and post-treatment tumor tissues, 28 patients (39%) were initially positive for MET expression, and 43 (61%) were negative. Twenty-five patients (35%) showed significant changes in MET expression after treatment (P=0.007). Additionally, there was a concomitant overexpression of MET and HER2 in a subset of GC patients. MET inhibitor volitinib could significantly inhibit cell proliferation and xenograft growth in vitro and in vivo in MKN45 cells with MET and phosphorylated MET (pMET) high expressions via suppressing downstream PI3K/Akt and MAPK signaling pathways. Furthermore, combination therapy targeting both MET and HER2 demonstrated a synergistic antitumor activity.ConclusionsMET expression is altered post chemotherapy and MET status should be evaluated in real-time. Both MET and pMET expressions might need to be considered for patients suitable for volitinib treatment.

We predict asset returns and measure risk premia using a prominent technique from artificial intelligence -- deep sequence modeling. Because asset returns often exhibit sequential dependence that may not be effectively captured by conventional time series models, sequence modeling offers a promising path with its data-driven approach and superior performance. In this paper, we first overview the development of deep sequence models, introduce their applications in asset pricing, and discuss their advantages and limitations. We then perform a comparative analysis of these methods using data on U.S. equities. We demonstrate how sequence modeling benefits investors in general through incorporating complex historical path dependence, and that Long- and Short-term Memory (LSTM) based models tend to have the best out-of-sample performance.

With the development of electronic health records (EHRs), prenatal care examination records have become available for developing automatic prediction or diagnosis approaches with machine learning methods. In this paper, we study how to effectively learn representations applied to various downstream tasks for EHR data. Although several methods have been proposed in this direction, they usually adapt classic sequential models to solve one specific diagnosis task or address unique EHR data issues. This makes it difficult to reuse these existing methods for the early diagnosis of pregnancy complications or provide a general solution to address the series of health problems caused by pregnancy complications. In this paper, we propose a novel model RAPT, which stands for RepresentAtion by Pre-training time-aware Transformer. To associate pre-training and EHR data, we design an architecture that is suitable for both modeling EHR data and pre-training, namely time-aware Transformer. To handle various characteristics in EHR data, such as insufficiency, we carefully devise three pre-training tasks to handle data insufficiency, data incompleteness and short sequence problems, namely similarity prediction, masked prediction and reasonability check. In this way, our representations can capture various EHR data characteristics. Extensive experimental results for four downstream tasks have shown the effectiveness of the proposed approach. We also introduce sensitivity analysis to interpret the model and design an interface to show results and interpretation for doctors. Finally, we implement a diagnosis system for pregnancy complications based on our pre-training model. Doctors and pregnant women can benefit from the diagnosis system in early diagnosis of pregnancy complications.

Purpose: The aim of this study was to investigate the role of BIRC5 for early diagnosis and prognosis in clear-cell renal cell carcinoma (ccRCC) by studying the expression of BIRC5 and the correlation between BIRC5 expression and clinicopathological parameters and prognosis in ccRCC. Methods: The BIRC5 expression in ccRCC tissues and normal kidney tissues was measured using the Cancer Genome Atlas database and the Human Protein Atlas database. The correlation between BIRC5 expression and clinicopathological parameters and prognosis in ccRCC was analyzed using UALCAN, the Kaplan-Meier plotter, GEPIA, and SurvExpress. Thirteen-paired ccRCC plasma samples were used to verify the BIRC5 early diagnosis value of ccRCC. Results: The BIRC5 expression is significantly higher in ccRCC than in normal kidney tissues, and is correlated with the clinical stage and pathological grade of ccRCC (p < 0.05). The result of analyzing the relationship between BIRC5 expression and outcomes in ccRCC indicates that a high BIRC5 expression is an independent prognostic factor affecting the overall survival and disease-free survival of ccRCC (p < 0.05). Compared with normal kidney tissues, the immunohistochemical test shows that BIRC5 is significantly upregulated in ccRCC tissues. mRNA expression levels of BIRC5 were significantly higher in the ccRCC plasma than normal (p < 0.05). Conclusions: The high expression of BIRC5 is an important indicator for the prognosis of ccRCC, which makes BIRC5 an effective biomarker for predicting the prognosis of patients in ccRCC. BIRC5 may be a great potential biomarker for early diagnosis of ccRCC.

Purpose: To develop and verify an early prediction model of gestational diabetes mellitus (GDM) using machine learning algorithm.Methods: The dataset collected from a pregnant cohort study in eastern China, from 2017 to 2019. It was randomly divided into 75% as the training dataset and 25% as the test dataset using the train_test_split function. Based on Python, four classic machine learning algorithm and a New-Stacking algorithm were first trained by the training dataset, and then verified by the test dataset. The four models were Logical Regression (LR), Random Forest (RT), Artificial Neural Network (ANN) and Support Vector Machine (SVM). The sensitivity, specificity, accuracy, and area under the Receiver Operating Characteristic Curve (AUC) were used to analyse the performance of models.Results: Valid information from a total of 2811 pregnant women were obtained. The accuracies of the models ranged from 80.09% to 86.91% (RF), sensitivities ranged from 63.30% to 81.65% (SVM), specificities ranged from 79.38% to 97.53% (RF), and AUCs ranged from 0.80 to 0.82 (New-Stacking).Conclusion: This paper successfully constructed a New-Stacking model theoretically, for its better performance in specificity, accuracy and AUC. But the SVM model got the highest sensitivity, the SVM model was recommends as the prediction model for clinical.

Afterpulsing is a critical non-ideal factor of SPAD at telecommunication wavelength, which limits its performance. In order to evaluate the effects of afterpulsing, a new bit error rate (BER) model of SPAD-based photon-counting communication system including afterpulsing is built. To verify the theoretical model, an experimental system is founded. Compared with experimental results, the new BER model is found to be in good agreement. Based on the new BER model, the error performance of photon-counting communication system is investigated. The results indicate that maintaining the afterpulsing probability (AP) of 0, 0.1 will bring 1.2 to 3-dB sensitivity benefit than AP of 0.25. Maintaining the total number of gates per bit above 20 can significantly improve the system sensitivity or background tolerance. Inhibiting the AP below 0.1 is critical for expanding the scale of SPADs array to enhance system error performance.

In order to detect weak signal photons from background radiation, using an array of InGaAs/InP single-photon avalanche diode (SPAD) as a detector is receiving increasing interest. Firstly, the canonical structure and operation principle of SPAD are analyzed. Then, based on the basic parameters (excess bias, operating temperature, absorption region thickness, and multiplication region thickness), the integrated SPAD model of performance parameters, which combines photon detection efficiency (PDE) and dark count rate (DCR), is built. Moreover, a scheme of utilizing SPADs array to detect weak signal photons and an accurate bit error rate (BER) model of photon-counting communication with the limitation of non-ideal factors of SPAD are studied. Then, the joint model of the integrated SPAD model and the BER model is founded. According to the joint model, the design considerations of SPAD to improve the BER in photon-counting communication are investigated. In addition, in regard to a specific photon-counting communication system, the basic parameters of SPAD can be effectively optimized through the joint model.

We develop hierarchical coordination frameworks to optimally manage active and reactive power dispatch of number of spatially distributed electric vehicles (EVs) incorporating distribution grid level constraints. The frameworks consist of detailed mathematical models, which can benefit the operation of both entities involved, i.e., the grid operations and EV charging. The first model comprises of a comprehensive optimal power flow model at the distribution grid level, while the second model represents detailed optimal EV charging with reactive power support to the grid. We demonstrate benefits of coordinated dispatch of active and reactive power from EVs using a 33-node distribution feeder with large number of EVs (more than 5000). Case studies demonstrate that, in constrained distribution grids, coordinated charging reduces the average cost of EV charging if the charging takes place at nonunity power factor mode compared to unity power factor. Similarly, the results also demonstrate that distribution grids can accommodate charging of increased number of EVs, if EV charging takes place at nonunity power factor mode compared to unity power factor.

BackgroundThe Cancer Genome Atlas (TCGA) project shed light on the vital role of tumor molecular features in predicting endometrial cancer patients’ prognosis. This study aims to investigate the survival impact of surgical approaches on patients with different genetic alterations.Methods473 endometrial cancer patients from TCGA database were selected. To analyze the prognostic impact of surgical approach, survival analyses were conducted in patients with different molecular features. Finally, a simplified molecular stratification model was established to select patients suitable for open or minimally invasive surgery (MIS).ResultsIn our cohort, 291 patients received open surgery and 182 received MIS. Molecular features influenced patients’ survival after different surgical approaches. Based on survival analyses, three molecular subtypes were generated, with subtype 1 harboring POLE mutation (POLEmt), microsatellite-instability high (MSI-H), homologous recombination repair (HRR) pathway mutation or MUC16 mutation (MUC16mt); subtype 3 carrying TP53 mutation; and subtype 2 without specific molecular feature. The survival influence of molecular subtypes depended on surgical approaches. In the open surgery cohort, three subtypes showed similar survival outcome, while in the MIS cohort, prognosis varied significantly among three subtypes, with subtype 1 the best and subtype 3 the worst. In stepwise Cox regression, molecular subtype was an independent predictor of recurrence-free survival in patients receiving MIS (p < 0.001).ConclusionThe molecular features of endometrial cancer are associated with patients’ prognosis after different surgical approaches. MIS should be recommended in patients with POLEmt, MSI-H, HRR pathway mutation or MUC16mt, while for patients with TP53 mutation, open surgery is better concerning oncological safety.

Objective: The aim of this study was to describe the clinicopathological features of EC patients with bone metastases treated with surgery and to systematically review the literature. Methods: We performed a retrospective study to include patients with bone metastases of EC at Peking University People’s Hospital during 2000 to 2019. Clinicopathological features and survival outcomes were collected.Results: Among the 1662 patients with EC, 14 (0.84%) were identified with bone metastases, and all were treated surgically. 13 cases were analyzed. Four had bone metastases when diagnosed; and the remaining 9 cases had bone metastases when first relapsed with the median time to recurrence of 13 months (range, 5-144). The median age of the 13 patients was 58 years old (range, 45-76). Twelve were endometrioid carcinoma. The majority sites of bone metastases were pelvis, followed by the spine. The median overall survival (OS) was 57 months. We further combined the 13 patients with another 24 cases identified from literature research. There was no significant difference in clinicopathological characteristics between the patients with bone metastases when diagnosed and when first relapsed. The median OS was numerically longer for patients with bone metastases when diagnosed than when first relapsed (57 vs. 36 months, P=0.084). The median OS was significantly longer in patients with endometrioid carcinoma than in those with type II carcinoma (57 vs. 4 months, P=0.003). Conclusions: Patients with bone metastases of EC might benefit from comprehensive treatment based on surgery, as symptoms can be palliated and survival can be extended probably.

We use proprietary data on intraday transactions at a futures brokerage to analyze the link between implied leverage, trading performance, and the sources of profits/losses, conditional on investor skill. We measure skill during a training period, and analyze performance out of sample. Levered positions stimulate de facto liquidity provision by skilled investors, who earn 19.3 bps per leverage unit. Unskilled investors' leverage amplifies their losses, particularly those stemming from gambling proclivity. Across all individuals and institutions, forced liquidations largely account for the negative impact of leverage on performance. Regulatory increases in required margins decrease (enhance) skilled (unskilled) investors' performance.

The impact of molecular alterations on programmed death-ligand 1 (PD-L1) combined positive score (CPS) is not well studied in gastroesophageal adenocarcinomas (GEAs). We aimed to characterize genomic features of tumors with different CPSs in GEAs.Genomic alterations of 2518 GEAs were compared in three groups (PD-L1 CPS ≥ 10, high; CPS = 1-9, intermediate; CPS1, low) using next-generation sequencing. We assessed the impact of gene mutations on the efficacy of immune checkpoint inhibitors (ICIs) and tumor immune environment based on the Memorial Sloan Kettering Cancer Center and The Cancer Genome Atlas databases.High, intermediate, and low CPSs were seen in 18%, 54% and 28% of GEAs, respectively. PD-L1 positivity was less prevalent in women and in tissues derived from metastatic sites. PD-L1 CPS was positively associated with mismatch repair deficiency/microsatellite instability-high, but independent of tumor mutation burden distribution. Tumors with mutations in KRAS, TP53, and RAS-mitogen-activated protein kinase (MAPK) pathway were associated with higher PD-L1 CPSs in the mismatch repair proficiency and microsatellite stability (pMMRMSS) subgroup. Patients with RAS-MAPK pathway alterations had longer overall survival (OS) from ICIs compared to wildtype (WT) patients [27 versus 13 months, hazard ratio (HR) = 0.36, 95% confidence interval (CI): 0.19-0.7, P = 0.016] and a similar trend was observed in the MSS subgroup (P = 0.11). In contrast, patients with TP53 mutations had worse OS from ICIs compared to TP53-WT patients in the MSS subgroup (5 versus 21 months, HR = 2.39, 95% CI: 1.24-4.61, P = 0.016).This is the largest study to investigate the distinct genomic landscapes of GEAs with different PD-L1 CPSs. Our data may provide novel insights for patient selection using mutations in TP53 and RAS-MAPK pathway and for the development of rational combination immunotherapies in GEAs.

Introduction/Background Recent studies and clinical trials demonstrated the vital significance of BRCA mutational status in ovarian cancer treatment, but related evidence in endometrial cancer (EC) is still limited. This study aims to investigate the role of BRCA mutations in predicting EC patients’ survival. Methodology 510 eligible cases from the Cancer Genome Atlas database were retrospectively analysed. Clinicopathological characteristics of patients with different BRCA1/2 mutational status were compared. To analyse the survival influence of BRCA1/2 mutation, Kaplan-Meier survival analyses and Cox regressions were conducted. In order to control confounding bias between groups, propensity score matching method was used. Results Among the eligible patients, 11 (2.2%) harboured BRCA1 mutations, 43 (8.4%) harboured BRCA2 mutations, and 36 (7.1%) harboured both. Body mass index, rates of hypertension history, proportion of non-endometrioid histology and rates of positive peritoneal cytology were lower in BRCA1/2 mutant patients compared with the wild-type counterpart (p = 0.020, 0.048, 0.001 and 0.012, respectively). Patients with BRCA1/2 mutations showed longer overall (OS) and recurrence-free survival (RFS) (in Kaplan-Meier analyses, p Conclusion BRCA1/2 mutations could predict better survival outcome in high-grade endometrioid EC patients, indicating the value of BRCA testing in EC clinical management. Disclosures This work was supported by the National Natural Science Foundation of China (81972426 and 81874108), Special Projects for Strengthening Basic Research of Peking University (BMU2018JC005), National Key Technology R&D Program of China (2019YFC1005200 and 2019YFC1005201). The authors have no potential conflict of interest to disclose.

Background The capacity of the liver to restore its architecture and function assures good prognoses of patients who suffer serious hepatic injury or cancer resection. In our previous study, we initially found that the P53/miR-34a/SIRT1 positive feedback loop has a remarkable negative regulatory effect, which is related to the termination of liver regeneration. Here, we described how P53/miR-34a/SIRT1 positive feedback loop controls liver regeneration and its possible relationship with liver cancer.Method We performed partial hepatectomy (PH) in mice transfected with adenovirus (Ade) overexpressing P53 and adenovirus-associated virus (AAV) knock-downing miR-34a. LR was analyzed by liver weight/body weight, serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels and cell proliferation, and the related cellular signals were investigated. Bile acid (BA) levels during LR were analyzed by metabolomics of bile acids. Results We found that the P53/miR-34a/SIRT1 positive feedback loop was activated in the late phase of LR. Overexpression of P53 terminated LR early and enhanced P53/miR-34a/SIRT1 positive feedback loop expression and its proapoptotic effect. Mice from the Ade-P53 group showed smaller livers and higher levels of serum ALT and AST than control mice. While knock-down of miR-34a abolished P53/miR-34a/SIRT1 positive feedback loop during LR. Mice from anti-miR-34a group showed larger livers and lower levels of PCNA-positive cells than control mice. T-β-MCA increased gradually during LR and peaked at 7 days after PH. T-β-MCA inhibited cell proliferation and promoted cell apoptosis via facilitating the P53/miR-34a/SIRT1 positive feedback loop during LR by suppressing FXR/SHP. Conclusion The P53/miR-34a/SIRT1 positive feedback loop plays an important role in the termination of LR. Our findings shed light on the molecular and metabolic mechanisms of LR termination and provide a potential therapeutic alternative for treating P53-wild-type HCC patients.

In the paper, aperture averaging of Modulating Retro-Reflector (MRR) optical communication is studied. The simulation model based on multi-phase screen method is established to analyze the scintillation index of MRR with different aperture. Also experiments are demonstrated to verify the simulation results. The results show that not only the aperture of receiver in interrogator but also the aperture of retro-reflector will affect the fluctuation of power, and these will be helpful to design the system of MRR.

Highly nonlinear photonic crystal fibers (PCFs) can be used as a kind of new ways to realize the supercontinuum generation because of their highly nonlinear effects. Based on a type of material-filled method and structure, novel highly nonlinear PCFs are proposed to compare with the conventional highly nonlinear fiber for supercontinuum generation in this manuscript. The nonlinear coefficient and supercontinuum spectra varying with different parameters such as the hole-diameter, the hole-pitch and the refractive index of the filled material are investigated respectively and analyzed numerically. Even if lower incident optical power is used on the designed PCFs than that on the conventional fiber and the previous proposed PCFs, broadband supercontinuum can still be obtained easily from highly nonlinear effects. The results demonstrate that it is possible for this kind of novel PCFs to achieve the maximum 450 nm supercontinuum spectra, even though it is only 50-cm-long fiber and pumped by substantially lower incident optical power of 1.5 W near the wavelength of 1.55 µm.

Introduction: Novel therapies are needed for the treatment of motion sickness given the inadequate relief and bothersome and dangerous adverse effects of currently approved therapies. Neurokinin-1 (NK1) receptor antagonists have the potential to be effective in improving the symptoms of motion sickness, given the involvement of Substance P in nauseogenic and emetic pathways and the expression of NK1 receptors in the gastrointestinal system. Here we evaluated the efficacy of tradipitant, a novel NK1 receptor antagonist, in preventing motion sickness in variable sea conditions. Methods: A total of 126 adults participated in the Motion Sifnos study. Groups of participants were assigned to one of seven boat trips lasting ~4 h on the Pacific Ocean. Participants were randomized 1:1 to tradipitant 170 mg or placebo and completed the Motion Sickness Severity Scale (MSSS) every 30 min, in addition to other assessments. Severity of motion sickness was assessed with the incidence of vomiting and the MSSS. Results: Participants on tradipitant had a significantly lower incidence of vomiting as compared to those on placebo across all boat trips (tradipitant = 17.5%, placebo = 39.7%, p = 0.0039). For trips exposed to rough sea conditions, the difference in the incidence of vomiting between the groups was more dramatic (tradipitant = 15.79%, placebo = 72.22%, p = 0.0009). Across these trips, motion sickness symptoms were significantly lower in the tradipitant group compared to the placebo group (tradipitant = 3.19, placebo = 4.57, p = 0.0235). Discussion: Tradipitant has the potential to be an effective therapy for the prevention of vomiting and treatment of nausea in people with motion sickness.

Purpose: Defective DNA damage response (DDR) is a hallmark of cancer leading to genomic instability and is associated with chemosensitivity. Although the mismatch repair system has been extensively studied, the clinical implications of other mechanisms associated with DDR alterations in patients with colorectal cancer remain unclear. This study aimed to understand DDR pathways alterations and their association with common clinical features in patients with colorectal cancer. Experimental Design: Next-generation sequencing and whole-transcriptome sequencing were conducted using formalin-fixed paraffin-embedded samples submitted to a commercial Clinical Laboratory Improvement Amendments–certified laboratory. Samples with pathogenic or presumed pathogenic mutations in 29 specific DDR-related genes were considered as DDR-mutant (DDR-MT) and the remaining samples as DDR-wild type (DDR-WT). Results: Of 9,321 patients with colorectal cancer, 1,290 (13.8%) were DDR-MT. The frequency of DDR-MT was significantly higher in microsatellite instability-high (MSI-H) cases than in microsatellite stable cases (76.4% vs. 9.5%). The DDR-MT genotype was higher in the right-sided, RAS-wild, BRAF-mutant, and CMS1 subgroups. However, these associations were primarily confounded by the distribution of MSI status. Compared with the DDR-WT tumors, the DDR-MT tumors had a higher mutational burden and gene expression levels in the immune-related pathway, which were independent of MSI status. Conclusions: We characterized a distinct subgroup of patients with colorectal cancer with tumors harboring mutations in the DDR-related genes. These patients more commonly had MSI-H tumors and exhibited an activated immune signature regardless of their tumor's MSI status. These findings warrant further investigations to develop personalized treatment strategies in this significant subgroup of patients with colorectal cancer.

International audience; Recent years witness the rapidly-growing business of ride-on-demand (RoD) services such as Uber, Lyft and Didi. Unlike taxi services, these emerging transportation services use dynamic pricing to manipulate the supply and demand, and to improve service responsiveness and quality. Despite this, on the drivers' side, dynamic pricing creates a new problem: how to seek for passengers in order to earn more under the new pricing scheme. Seeking strategies have been studied extensively in traditional taxi service, but in RoD service such studies are still rare and require the consideration of more factors such as dynamic prices, the status of other transportation services, etc. In this paper, we develop ROD-Revenue, aiming to mine the relationship between driver revenue and factors relevant to seeking strategies, and to predict driver revenue given features extracted from multi-source urban data. We extract basic features from multiple datasets, including RoD service, taxi service, POI information, and the availability of public transportation services, and then construct composite features from basic features in a product-form. The desired relationship is learned from a linear regression model with basic features and high-dimensional composite features. The linear model is chosen for its interpretability-to quantitatively explain the desired relationship. Finally we evaluate our model by predicting drivers' revenue. We hope that ROD-Revenue not only serves as an initial analysis of seeking strategies in RoD service, but also helps increasing drivers' revenue by offering useful guidance.

We have designed and developed the central network management architecture system of the network intelligent experimental system, the network intelligent experimental equipment and system of various courses, which integrates modern information technology, database management technology, network technology and experimental teaching, truly embodies the network and intelligent management function of the experimental teaching point, and makes the management level and teaching of the experimental teaching conditions, teaching quality and openness will be greatly improved. At the same time, the system supports secondary development, allows teachers to add new experimental projects according to their own needs, and realizes the intelligent management of the network. This study can provide insights for the students and teachers on how to use Network Intelligent Experiment System in daily work.