Evolutionary Multi-Tasking Optimization for High-Efficiency Time Series Data Clustering

Time series clustering is a challenging problem due to the large-volume,high-dimensional,and warping characteristics of time series data.Traditional clustering methods often use a single criterion or distance measure,which may not capture all the features of the data.This paper proposes a novel method for time series clustering based on evolutionary multi-tasking optimization,termed i-MFEA,which uses an improved multifactorial evolutionary algorithm to optimize multiple clustering tasks simultaneously,each with a different validity index or distance measure.Therefore,i-MFEA can produce diverse and robust clustering solutions that satisfy various preferences of decision-makers.Experiments on two artificial datasets show that i-MFEA outperforms single-objective evolutionary algorithms and traditional clustering methods in terms of convergence speed and clustering quality.The paper also discusses how i-MFEA can address two long-standing issues in time series clustering:the choice of appropriate similarity measure and the number of clusters.

Massive feature extraction for explaining and foretelling hydroclimatic time series forecastability at the global scale

Statistical analyses and descriptive characterizations are sometimes assumed to be offering information on time series forecastability.Despite the scientific interest suggested by such assumptions,the relationships between descriptive time series features(e.g.,temporal dependence,entropy,seasonality,trend and linearity features)and actual time series forecastability(quantified by issuing and assessing forecasts for the past)are scarcely studied and quantified in the literature.In this work,we aim to fill in this gap by investigating such relationships,and the way that they can be exploited for understanding hydroclimatic forecastability and its patterns.To this end,we follow a systematic framework bringing together a variety of–mostly new for hydrology–concepts and methods,including 57 descriptive features and nine seasonal time series forecasting methods(i.e.,one simple,five exponential smoothing,two state space and one automated autoregressive fractionally integrated moving average methods).We apply this framework to three global datasets originating from the larger Global Historical Climatology Network(GHCN)and Global Streamflow Indices and Metadata(GSIM)archives.As these datasets comprise over 13,000 monthly temperature,precipitation and river flow time series from several continents and hydroclimatic regimes,they allow us to provide trustable characterizations and interpretations of 12-month ahead hydroclimatic forecastability at the global scale.We first find that the exponential smoothing and state space methods for time series forecasting are rather equally efficient in identifying an upper limit of this forecastability in terms of Nash-Sutcliffe efficiency,while the simple method is shown to be mostly useful in identifying its lower limit.We then demonstrate that the assessed forecastability is strongly related to several descriptive features,including seasonality,entropy,(partial)autocorrelation,stability,(non)linearity,spikiness and heterogeneity features,among others.We further(i)show that,if such descriptive information is available for a monthly hydroclimatic time series,we can even foretell the quality of its future forecasts with a considerable degree of confidence,and(ii)rank the features according to their efficiency in explaining and foretelling forecastability.We believe that the obtained rankings are of key importance for understanding forecastability.Spatial forecastability patterns are also revealed through our experiments,with East Asia(Europe)being characterized by larger(smaller)monthly temperature time series forecastability and the Indian subcontinent(Australia)being characterized by larger(smaller)monthly precipitation time series forecastability,compared to other continental-scale regions,and less notable differences characterizing monthly river flow from continent to continent.A comprehensive interpretation of such patters through massive feature extraction and feature-based time series clustering is shown to be possible.Indeed,continental-scale regions characterized by different degrees of forecastability are also attributed to different clusters or mixtures of clusters(because of their essential differences in terms of descriptive features).

Large-Scale KPI Anomaly Detection Based on Ensemble Learning and Clustering

Anomaly detection using KPI(Key Performance Indicator)is critical for Internet-based services to maintain high service availability.However,given the velocity,volume,and diversified nature of monitoring data,it is difficult to obtain enough labelled data to build an accurate anomaly detection model for using supervised machine leaning methods.In this paper,we propose an automatic and generic transfer learning strategy:Detecting anomalies on a new KPI by using pretrained model on existing selected labelled KPI.Our approach,called KADT(KPI Anomaly Detection based on Transfer Learning),integrates KPI clustering and model pretrained techniques.KPI clustering is used to obtain the similarity of different KPI data's distribution,and applied transfer knowledge from source dataset to the target dataset by model pretrained technique.In our evaluation using real-world KPIs from large Internet-based services,the clustering algorithm used to detect various KPI curve pattern achieve the best classification effect and accuracy More importantly,further evaluation on 30 KPIs shows that KADT can significantly reduce the time overhead of the model training with little loss of accuracy.

Bioinformatics Analysis of Genes and Pathways of CD11b^＋/Ly6C^intermediate Macrophages after Renal Ischemia-Reperfusion Injury

Renal ischemia-reperfusion injury（IRI） is a major cause of acute kidney injury（AKI）,which could induce the poor prognosis.The purpose of this study was to characterize the molecular mechanism of the functional changes of CD11 b＋/Ly6 Cintermediate macrophages after renal IRI.The gene expression profiles of CD11 b＋/Ly6 Cintermediate macrophages of the sham surgery mice,and the mice 4 h,24 h and 9 days after renal IRI were downloaded from the Gene Expression Omnibus database.Analysis of m RNA expression profiles was conducted to identify differentially expressed genes（DEGs）,biological processes and pathways by the series test of cluster.Protein-protein interaction network was constructed and analysed to discover the key genes.A total of 6738 DEGs were identified and assigned to 20 model profiles.DEGs in profile 13 were one of the predominant expression profiles,which are involved in immune cell chemotaxis and proliferation.Signet analysis showed that Atp5 a1,Atp5 o,Cox4 i,Cdc42,Rac2 and Nhp2 were the key genes involved in oxidation-reduction,apoptosis,migration,M1-M2 differentiation,and proliferation of macrophages.RPS18 may be an appreciate reference gene as it was stable in macrophages.The identified DEGs and their enriched pathways investigate factors that may participate in the functional changes of CD11 b＋/Ly6 Cintermediate macrophages after renal IRI.Moreover,the vital gene Nhp2 may involve the polarization of macrophages,which may be a new target to affect the process of AKI.

Construction of Regional Economic Vitality Model

Regional economic vitality reflects the scale and development potential of a region’s economy.It largely determines the development of the city,and is also affected by many factors such as population competitiveness,corporate competitiveness,market vitality,innovation vitality,and environmental vitality.A pilot model was constructed with Hebei Province as the inspection area.Quantitative measurement of regional economic vitality was made by finding 21 indicators that indirectly or indirectly affect the economic vitality of Hebei Province.By analyzing the data of 21 indicators for nearly 10 years,the time series clustering is used to achieve the dimensionality reduction of the indicators.After the dimension reduction,it is divided into four categories:overall scale,development potential,market vitality,and innovation vitality.Construct the economic vitality structure model of Hebei Province,and determine the four types of contribution to economic vitality and compare them.On this basis,more accurately grasp the indicators that affect economic vitality and work out reasonable and effective action plans.From the perspective of human resources and corporate vitality,analyze how the action plan accurately affects the economic vitality of Hebei Province[1].The 11 cities in Hebei Province are the target of regional economic vitality.The economic vitality structure model constructed uses the required contribution value to select priority indicators.Finally,the six indicators of GPD,GPD growth rate,fiscal revenue,fiscal revenue growth rate,number of industrial enterprises above designated size,and total profit of industrial enterprises above designated size were established for eleven cities in Hebei Province to construct a TOPSIS scoring model,and calculation rankings were conducted through MATLAB.Results The top three cities were Shijiazhuang,Tangshan and Cangzhou.

Spatio-temporal characteristics of human activities using location big data in Qilian Mountain National Park

Human activities significantly impact the environment.Understanding the patterns and distribution of these activities is crucial for ecological protection.With location-based technology advancement,big data such as location and trajectory data can be used to analyze human activities on finer temporal and spatial scales than traditional remote sensing data.In this study,Qilian Mountain National Park(QMNP)was chosen as the research area,and Tencent location data were used to construct time series data.Time series clustering and decomposition were performed,and the spatio-temporal distribution characteristics of human activities in the study area were analyzed in conjunction with GPS trajectory data and land use data.The study found two distinct human activity patterns,Pattern A and Pattern B,in QMNP.Compared to Pattern B,Pattern A had a higher volume of location data and clear nighttime peaks.By incorporating land use and trajectory data,we conclude that Pattern A and Pattern B represent the activity patterns of the resident and tourist populations,respectively.Moreover,the study identified seasonal variations in human activities,with human activity in summer being approximately two hours longer than in winter.We also conducted an analysis of human activities in different counties within the study area.