Affinity propagation (AP) was recently introduced as an unsupervised learning algorithm for exemplar based clustering. In this paper novel text document clustering algorithm has been developed based on vector space model, phrases and affinity propagation clustering algorithm. Proposed algorithm can be called Phrase affinity clustering (PAC). PAC first finds the phrase by ukkonen suffix tree construction algorithm, second finds the vector space model using tf-idf weighting scheme of phrase. Third calculate the similarity matrix form VSD using cosine similarity .In Last affinity propagation algorithm generate the clusters .F-Measure ,Purity and Entropy of Proposed algorithm is better than GAHC ,ST-GAHC and ST-KNN on OHSUMED ,RCV1 and News group data sets.
Mobile ad hoc networks are comprised of nodes that must cooperate to dynamically establish routes using wireless links. Routes may involve multiple hops with each node acting as a host and router. Since ad hoc networks typically work in an open un-trusted environment with little physical security, they are subject to a number of unique security attacks like wormhole attack. The wormhole attack is considered to be a serious security attack in multi-hop ad hoc networks. In wormhole attack, attacker makes tunnel from one end of the network to the other, nodes stay in different location on two ends of tunnel believe that they are true neighbours and makes conversation through the wormhole link. Unlike many other attacks on ad-hoc routing, a wormhole attack cannot be prevented with cryptographic solutions because intruders neither generate new, nor modify existing, packets, but rather forward existing ones. In this paper a simple technique to effectively detect wormhole attacks without the need for special hardware and/or strict location or synchronization requirements is proposed. The proposed technique makes use of variance in routing information between neighbours to detect wormholes. The base of dissertation is to find alternative path from source to second hop and calculate the number of hops to detect the wormhole.
Abstract-Mobile ad-hoc communication is a spontaneous network because the topology is not stationary but self-organized. This requires that during the time MANET it operational, all the processes regarding discovering the topology, delivery of data packets and internal management communications must be taken care by the node(s) themselves. This implies the criteria for selection of Cluster Head (CH) and the routing related protocols are to be integrated into mobile node(s).The very facts that MANET is challenging and innovative areas of wireless networks, makes it more vulnerable in term of routing and flooding attacks. In this paper, a node trust calculation methodology is proposed which calculate the trust value of each node and applies fuzzy logic to detect wormhole, Black-hole (Routing attack) and distributed denial of service attack (DDOS/Flooding) in dynamic environment.
Abstract-Past observations have shown that a frequent item set mining algorithm are purported to mine the closed ones because the finish provides a compact and a whole progress set and higher potency. Anyhow, the newest closed item set mining algorithms works with candidate maintenance combined with check paradigm that is pricey in runtime yet as space usage when support threshold is a smaller amount or the item sets gets long. Here, we show, CEG&REP that could be a capable algorithm used for mining closed sequences while not candidate. It implements a completely unique sequence finality verification model by constructing a Graph structure that build by an approach labeled "Concurrent Edge Prevision and Rear Edge Pruning" briefly will refer as CEG&REP. a whole observation having sparse and dense real-life knowledge sets proved that CEG&REP performs bigger compared to older algorithms because it takes low memory and is quicker than any algorithms those cited in literature frequently.
The present work deals with a new symmetric key cryptographic method using dynamic key. The demand for adequate security to electronic data system grows high over the decades. In the present work the authors have used the Linear Congruential Generator (LCG) for generating key. This is a block cipher technique. The advantage of the present method is that for every pair of encryption & decryption operation a new dynamic key is generated thus the process is very hard to break.The cryptography no longer relies on long term shared keys which are vulnerable under cryptanalysis attacks. It is impossible to detect patterns with which to perform cryptanalysis on the dynamic key.In the present work the authors have introduced concept of dynamic key with symmetric cryptography. Dynamic key is similar to one time pad. In this paper, a dynamic key theory is described and mathematically analyzed.In the present method author proposed a cryptography system in which four rounds of encryption & decryption are performed. In each round different parts of dynamic key are applied in order to make it hard against cryptanalysis attacks.
Research Objective Heart failure (HF)–related hospitalizations are a growing public health burden, especially among older adults. Risk calculators for HF readmissions identify patients at high risk for readmissions who may benefit from outpatient interventions to improve outcomes and prevent readmissions. We hypothesized that incorporating additional variables and using machine learning approaches would improve the performance of these existing 30‐day readmission risk calculators. Study Design We evaluated the performance of several published risk calculators for predicting 30‐day readmission after heart failure hospitalizations: 1) using updating the coefficients based on data from a new, ethnically diverse HF population; 2) developing a new model that incorporates additional variables in addition to updated coefficients; and 3) developing new models with all variables using machine learning approaches. We used an 80%/20% split sampling for development and validation testing. The risk calculators tested included the LACE+ Index and Yale CORE, which included traditional clinical variables such as comorbidities, laboratory values, prior utilization, and vital signs. For the model with additional variables, we included all variables used in the original models with the addition of Comorbidity Point Score (COPS), Laboratory‐based Acute Physiology Score (LAPS), cardiovascular medications, discharge status, and socioeconomic status. We evaluated the original model and the original plus model, the additional variables using logistic regression. For the machine learning approaches, we used lasso penalized regression and gradient boosting with k‐fold cross‐validation to avoid overfitting. We assessed model performance using area under the curve (AUC) and calibration plots. Population Studied We identified 38,234 adults hospitalized for HF between 2012 and 2017 within Kaiser Permanente Northern California, an integrated health care delivery system covering over 4.4 million members. Principal Findings Discrimination (AUC) was poor using original models, LACE+ [0.60 (0.59‐0.62)] and Yale CORE [0.57 (0.56‐0.59)]. Including the additional variables resulted in a small improvement in AUC: LACE+ [0.62 (0.60‐0.64)] and Yale CORE [0.62 (0.60‐0.64)]. The lasso model [0.67 (0.65, 0.68) and gradient boosting model [0.67 (0.65‐0.68)] resulted in greater improvement. Calibration plots showed generally good calibration across all models with modest improvements after adding the additional data domains and using the machine learning approaches. Conclusions Incorporating additional data domains led to small, statistically significant improvements in model discrimination while maintaining good calibration for published models to predict readmission. Machine learning approaches resulted in even greater improvement and overall moderate discrimination. Implications for Policy or Practice We were able to increase the utility of these published risk calculators for readmission after discharge from a HF hospitalization by including additional d...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.