: This thesis applies neural network feature selection techniques to multivariate time series data to improve prediction of a target time series. Two approaches to feature selection are used. First, a subset enumeration method is used to determine which financial indicators are most useful for aiding in prediction of the S&P 500 futures daily price. The candidate indicators evaluated include RSI, Stochastics and several moving averages. Results indicate that the Stochastics and RSI indicators result in better prediction results than the moving averages. The second approach to feature selection is calculation of individual saliency metrics. A new decision boundary-based individual saliency metric, and a classifier independent saliency metric are developed and tested. Ruck's saliency metric, the decision boundary based saliency metric, and the classifier independent saliency metric are compared for a data set consisting of the RSI and Stochastics indicators as well as delayed closing price values. The decision based metric and the Ruck metric results are similar, but the classifier independent metric agrees with neither of the other metrics. The nine most salient features, determined by the decision boundary based metric, are used to train a neural network and the results are presented and compared to other published results. (AN)

/pdf/nonlinear-time-series-analysis-1j3a35n4y3.pdf

Nonlinear Time Series Analysis.

Many ``real-world'' networks are clearly defined while most ``social'' networks are to some extent subjective. Indeed, the accuracy of empirically-determined social networks is a question of some concern because individuals may have distinct perceptions of what constitutes a social link. One unambiguous type of connection is sexual contact. Here we analyze data on the sexual behavior of a random sample of individuals, and find that the cumulative distributions of the number of sexual partners during the twelve months prior to the survey decays as a power law with similar exponents $\alpha \approx 2.4$ for females and males. The scale-free nature of the web of human sexual contacts suggests that strategic interventions aimed at preventing the spread of sexually-transmitted diseases may be the most efficient approach.

The Web of Human Sexual Contacts

応用水文 = Applied hydrology

This paper presents a survey of simulation and optimization modeling approaches used in reservoir systems operation problems. Optimization methods have been proved of much importance when used with simulation modeling and the two approaches when combined give the best results. The main objective of this review article is to discuss simulation, optimization and combined simulation–optimization modeling approach and to provide an overview of their applications reported in literature. In addition to classical optimization techniques, application and scope of computational intelligence techniques, such as, evolutionary computations, fuzzy set theory and artificial neural networks, in reservoir system operation studies are reviewed. Conclusions and suggestive remarks based on this survey are outlined, which could be helpful for future research and for system managers to decide appropriate methodology for application to their systems.

/pdf/simulation-optimization-modeling-a-survey-and-potential-1xvkgabzfk.pdf

Simulation-optimization modeling: a survey and potential application in reservoir systems operation.

Linear Theory of Hydrologic Systems

In the present study, the impact of land use–land cover (LULC) change and urbanization on floods are investigated for an expanding urban catchment of the Oshiwara River in Mumbai, India. For the study area, the land use change was estimated between 1966, 2001 and 2009 by using the topographic map and satellite images. The analysis of LULC change revealed that the change between 1966 and 2001 was slower than that between 2001 and 2009. The LULC analysis revealed a 74.84% increase in the built up area with a 42.8% decrease in open spaces between the years 1966 and 2009, with substantial increase in urbanization. The impact of LULC on flood hydrograph for different return periods was ascertained by using the HEC-GeoHMS and HEC-HMS models. In the past 43 years, the increase in peak runoff and runoff volume is marginally varied by 3.0% and 4.45% for the 100-year return period and 10.4% and 12.2% for the 2-year return period respectively, although the built-up area increased by 74.84%. The flood inundation area is increased by 5.61% for the 100-year return period and 6.04% for the 10-year return period between the same time period. The results showed that lower return periods led to a maximum change in peak discharge/volume of runoff compared to higher return periods for change in land use conditions. Further, a flood hazard analysis has been carried out and it showed that the area in highly hazardous zone is increased by 64.29% as compared to less hazardous zone where it is decreased by 32.14%. Overall, the total flood hazard area is increased by 22.27%. The developed flood plain and flood hazard maps can be used by the local Municipal body to prepare flood mitigation and early evacuation management plans during floods and as a criteria for insurance of any property by insurance organizations.

Impacts of land use–land cover change and urbanization on flooding: A case study of Oshiwara River Basin in Mumbai, India

To obtain optimal operating rules for storage reservoirs, large numbers of simulation and optimization models have been developed over the past several decades, which vary significantly in their mechanisms and applications. As every model has its own limitations, the selection of appropriate model for derivation of reservoir operating rule curves is difficult and most often there is a scope for further improvement as the model selection depends on data available. Hence, evaluation and modifications related to the reservoir operation remain classical. In the present study a Genetic Algorithm model has been developed and applied to Pechiparai reservoir in Tamil Nadu, India to derive the optimal operational strategies. The objective function is set to minimize the annual sum of squared deviation form desired irrigation release and desired storage volume. The decision variables are release for irrigation and other demands (industrial and municipal demands), from the reservoir. Since the rule curves are derived through random search it is found that the releases are same as that of demand requirements. Hence based on the present case study it is concluded that GA model could perform better if applied in real world operation of the reservoir.

Single Reservoir Operating Policies Using Genetic Algorithm

Land use–land cover (LULC) change in space and time is the main cause behind the changing hydrological processes, ecosystem and environment in urban catchments. In the present study, the main focus was on evaluation of spatial and temporal variation of land use and land cover change in a major coastal urban catchment of Mumbai City, India, called Mithi River catchment. The LULC is derived from the topographic map surveyed in the year 1966 and satellite image for the year 2009. The analysis from toposheet and remote sensing data showed that there is a rise in the built-up area of Mithi River catchment, Mumbai by 59.66 % between 1966 and 2009. It also showed adverse human-induced influences on the Mithi River course and its catchment. Flood hydrographs for different land use conditions were derived by using Soil Conservation Service-Curve Number hydrological model and kinematic wave model, for routing available within the HEC-HMS software. Flood plain maps as well as flood hazard maps for the different land use scenarios have been developed by integrating the models HEC-HMS and HEC-RAS with HEC-GeoHMS and HEC-GeoRAS as well as with GIS and remote sensing. Results obtained from the present study revealed marginal increases in the runoff peak discharges and volumes within the catchment. Even though the runoff change is marginal, combined with tidal influence, it may cause major flooding problem. The integrated modeling approach has been found to be very effective for flood estimation, flood plain and flood hazard mapping. The flood plain and flood hazard maps derived can be used as effective tool to minimize the damages within the flood-prone areas of the river basin for the Mumbai City.

Impacts of urbanization on flooding of a coastal urban catchment: a case study of Mumbai City, India

Multi-time-step ahead daily and hourly intermittent reservoir inflow prediction by artificial intelligent techniques using lumped and distributed data

Flooding in urban area is a major natural hazard causing loss of life and damage to property and infrastructure. The major causes of urban floods include increase in precipitation due to climate change effect, drastic change in land use–land cover (LULC) and related hydrological impacts. In this study, the change in LULC between the years 1966 and 2009 is estimated from the toposheets and satellite images for the catchment of Poisar River in Mumbai, India. The delineated catchment area of the Poisar River is 20.19 km2. For the study area, there is an increase in built-up area from 16.64 to 44.08% and reduction in open space from 43.09 to 7.38% with reference to total catchment area between the years 1966 and 2009. For the flood assessment, an integrated approach of Hydrological Engineering Centre-Hydrological Modeling System (HEC-HMS), HEC-GeoHMS and HEC-River analysis system (HEC-RAS) with HEC-GeoRAS has been used. These models are integrated with geographic information system (GIS) and remote sensing data to develop a regional model for the estimation of flood plain extent and flood hazard analysis. The impact of LULC change and effects of detention ponds on surface runoff as well as flood plain extent for different return periods have been analyzed, and flood plain maps are developed. From the analysis, it is observed that there is an increase in peak discharge from 2.6 to 20.9% for LULC change between the years 1966 and 2009 for the return periods of 200, 100, 50, 25, 10 and 2 years. For the LULC of year 2009, there is a decrease in peak discharge from 10.7% for 2-year return period to 34.5% for 200-year return period due to provision of detention ponds. There is also an increase in flood plain extent from 14.22 to 42.5% for return periods of 10, 25, 50 and 100 years for LULC change between the year 1966 and year 2009. There is decrease in flood extent from 4.5% for 25-year return period to 7.7% for 100-year return period and decrease in total flood hazard area by 14.9% due to provisions of detention pond for LULC of year 2009. The results indicate that for low return period rainfall events, the hydrological impacts are higher due to geographic characteristics of the region. The provision of detention ponds reduces the peak discharge as well as the extent of the flooded area, flood depth and flood hazard considerably. The flood plain maps and flood hazard maps generated in this study can be used by the Municipal Corporation for flood disaster and mitigation planning. The integration of available software models with GIS and remote sensing proves to be very effective for flood disaster and mitigation management planning and measures.

Vinayakam Jothiprakash

Papers

Impacts of land use–land cover change and urbanization on flooding: A case study of Oshiwara River Basin in Mumbai, India

Single Reservoir Operating Policies Using Genetic Algorithm

Impacts of urbanization on flooding of a coastal urban catchment: a case study of Mumbai City, India

Multi-time-step ahead daily and hourly intermittent reservoir inflow prediction by artificial intelligent techniques using lumped and distributed data

Hydrological impacts of land use–land cover change and detention basins on urban flood hazard: a case study of Poisar River basin, Mumbai, India