Best-first Search Research Articles

Efficient methods for finding an optimal network location for travel planning

In this paper, we describe an optimal network location for travel planning (ONLTP) query, a type of optimal location query. In trip planning, finding the optimal point for a group of users is a fundamental problem in spatial group query processing. Many previous studies have considered the problem of finding the optimal point. However, their queries using an exact method perform efficiently only when the users are closely distributed, not spread out in large road networks. In contrast, approximation methods use two different spatial indices, but they cannot control the trade-off between query performance and accuracy. We propose a method using G-trees [1, 2] to remedy these drawbacks. Our exact method is a concrete implementation of the best-first search in G-trees, and our approximation method further reduces the visited nodes of the exact method.

A Hybrid Mechanism to Detect DDoS Attacks in Software Defined Networks

DDoS (Distributed Denial-of-Service) attacks are among the cyberattacks that are increasing day by day and have caused problems for computer network servers. With the advent of SDN networks, they are not immune to these attacks, and due to the software-centric nature of these networks, this type of attack can be much more difficult for them, ignoring effective parameters such as port and Source IP in detecting attacks, providing costly solutions which are effective in increasing CPU load, and low accuracy in detecting attacks are of the problems of previously presented methods in detecting DDoS attacks. Given the importance of this issue,the purpose of this paper is to increase the accuracy of DDoS attack detection using the second order correlation coefficient technique based on ∅-entropy according to source IP and selection of optimal features.To select the best features, by examining the types of feature selection algorithms and search methods, the WrapperSubsetEval feature selection algorithm, the BestFirst search method, and the best effective features were selected. This study was performed on CTU-13 and ISOT datasets and the results were compared with other methods. The accuracy of the detection in this work indicates the high efficiency of the proposed approach compared to other similar methods.  

Some new perspectives for solving 0–1 integer programming problems using Balas method

Egon Balas’s additive algorithm, also known as implicit enumeration, is a technique that uses a branch-and-bound (B&B) approach to finding optimal solutions to 0–1 integer programming problems. Three common search strategies in B&B are depth-first search, breadth-first search and best-first search. The B&B approach generates a list of pending nodes to be evaluated and storage of these nodes becomes a memory issue for larger problems. In this paper, we propose a simple bookkeeping method that tracks the state of the problem using a single array when performing a depth-first search, dramatically reducing memory requirements. The method also provides the ability to calculate, at any point of the search, the theoretical maximum number of remaining nodes to be evaluated. We note in this paper that when using the best-first search strategy, the first candidate solution found is the optimal solution.

Improving software effort estimation using bio-inspired algorithms to select relevant features: An empirical study

Bio-inspired feature selection algorithms got the attention of the researchers in the domain of Software Development Effort Estimations (SDEE) because they can improve the prediction accuracy of existing estimation techniques, such as machine learning methods. This paper aims to analyze different feature selection algorithms and assess the role they can play to increase the accuracy of software development effort predictions. We have performed an empirical study considering commonly used bio-inspired feature selection algorithms in the domain of SDEE, i.e., Genetic Algorithm (GA), Particle Swarm Optimization, Ant Colony Optimization, Tabu Search, Harmony Search (HS), and Firefly algorithm, and four traditional non-bio-inspired algorithms, i.e., Best-First Search (BFS), Greedy Stepwise, Subset Forward Selection, and Random Search, used in combination with five widely used estimation techniques and applied to eight widely used SDEE datasets. The performed analysis suggests that almost all (bio-inspired) feature selection algorithms have outperformed the baseline estimation techniques (i.e., techniques employed without any feature selection algorithms) in the majority of the experiments and hence we can conclude that feature selection algorithms can help in the domain of SDEE to increase the prediction accuracy. Similarly, HS and GA are considered as best performed bio-inspired algorithms because they provided significantly better results than the non-bio-inspired algorithms in a greater number of experiments. Moreover, we also compared the results of various employed bio-inspired algorithms, and, again, GA and HS came out as the best performed bio-inspired feature selection algorithms. From our results, if we have to pick feature selection algorithms (from both bio- and non-bio-inspired) and recommend them for future investigations, we would suggest HS because it provided better effort predictions in more combinations of datasets and estimation techniques than the other considered bio- and non-bio-inspired algorithms. Among the non-bio-inspired algorithms, BFS is the one that provided better predictions. • Feature selection algorithms should be used to build effort estimation models. • Bio-inspired algorithms resulted better than non bio-inspired algorithms • Harmony Search and Genetic Algorithm performed better than other algorithms.

Mathematical modeling and two efficient branch and bound algorithms for job shop scheduling problem followed by an assembly stage

Purpose Two-stage production systems including a processing shop and an assembly stage are widely used in various manufacturing industries. These two stages are usually studied independently which may not lead to ideal results. This paper aims to deal with a two-stage production system including a job shop and an assembly stage. Design/methodology/approach Some exact methods are proposed based on branch and bound (B&B) approach to minimize the total completion time of products. As B&B approaches are usually time-consuming, three efficient lower bounds are developed for the problem and variable neighborhood search is used to provide proper upper bound of the solution in each branch. In addition, to create branches and search new nodes, two strategies are applied including the best-first search and the depth-first search (DFS). Another feature of the proposed algorithms is that the search space is reduced by releasing the precedence constraint. In this case, the problem becomes equivalent to a parallel machine scheduling problem, and the redundant branches that do not consider the precedence constraint are removed. Therefore, the number of nodes and computational time are significantly reduced without eliminating the optimal solution. Findings Some numerical examples are used to evaluate the performance of the proposed methods. Comparison result to mathematical model (mixed-integer linear programming) validates the performance accuracy and efficiency of the proposed methods. In addition, computational results indicate the superiority of the DFS strategy with regard to CPU time. Originality/value Studies about the scheduling problems for two-stage production systems including job shop followed by an assembly stage traditionally present approximate method and metaheuristic algorithms to solve the problem. This is the first study that introduces exact methods based on (B&B) approach.

AutoML for Multi-Label Classification: Overview and Empirical Evaluation.

Automated machine learning (AutoML) supports the algorithmic construction and data-specific customization of machine learning pipelines, including the selection, combination, and parametrization of machine learning algorithms as main constituents. Generally speaking, AutoML approaches comprise two major components: a search space model and an optimizer for traversing the space. Recent approaches have shown impressive results in the realm of supervised learning, most notably (single-label) classification (SLC). Moreover, first attempts at extending these approaches towards multi-label classification (MLC) have been made. While the space of candidate pipelines is already huge in SLC, the complexity of the search space is raised to an even higher power in MLC. One may wonder, therefore, whether and to what extent optimizers established for SLC can scale to this increased complexity, and how they compare to each other. This paper makes the following contributions: First, we survey existing approaches to AutoML for MLC. Second, we augment these approaches with optimizers not previously tried for MLC. Third, we propose a benchmarking framework that supports a fair and systematic comparison. Fourth, we conduct an extensive experimental study, evaluating the methods on a suite of MLC problems. We find a grammar-based best-first search to compare favorably to other optimizers.

Minimum-Delay Routing for Integrated Aeronautical Ad Hoc Networks Relying on Real Flight Data in the North-Atlantic Region

Relying on multi-hop communication techniques, aeronautical ad hoc networks (AANETs) seamlessly integrate ground base stations (BSs) and satellites into aircraft communications for enhancing the on-demand connectivity of planes in the air. The goal of the paper is to assess the performance of the classic shortest-path routing algorithm in the context of the real flight data collected in the North-Atlantic Region. Specifically, in this integrated AANET context we investigate the shortest-path routing problem with the objective of minimizing the total delay of the in-flight connection from the ground BS subject to certain minimum-rate constraints for all selected links in support of low-latency and high-speed services. Inspired by the best-first search and priority queue concepts, we model the problem formulated by a weighted digraph and find the optimal route based on the shortest-path algorithm. Our simulation results demonstrate that aircraft-aided multi-hop communications are capable of reducing the total delay of satellite communications, when relying on real historical flight data.

Best-First Beam Search

Decoding for many NLP tasks requires an effective heuristic algorithm for approximating exact search because the problem of searching the full output space is often intractable, or impractical in many settings. The default algorithm for this job is beam search—a pruned version of breadth-first search. Quite surprisingly, beam search often returns better results than exact inference due to beneficial search bias for NLP tasks. In this work, we show that the standard implementation of beam search can be made up to 10x faster in practice. Our method assumes that the scoring function is monotonic in the sequence length, which allows us to safely prune hypotheses that cannot be in the final set of hypotheses early on. We devise effective monotonic approximations to popular nonmonontic scoring functions, including length normalization and mutual information decoding. Lastly, we propose a memory-reduced variant of best-first beam search, which has a similar beneficial search bias in terms of downstream performance, but runs in a fraction of the time.

Performance Evaluation of a Proposed Machine Learning Model for Chronic Disease Datasets Using an Integrated Attribute Evaluator and an Improved Decision Tree Classifier

There is a consistent rise in chronic diseases worldwide. These diseases decrease immunity and the quality of daily life. The treatment of these disorders is a challenging task for medical professionals. Dimensionality reduction techniques make it possible to handle big data samples, providing decision support in relation to chronic diseases. These datasets contain a series of symptoms that are used in disease prediction. The presence of redundant and irrelevant symptoms in the datasets should be identified and removed using feature selection techniques to improve classification accuracy. Therefore, the main contribution of this paper is a comparative analysis of the impact of wrapper and filter selection methods on classification performance. The filter methods that have been considered include the Correlation Feature Selection (CFS) method, the Information Gain (IG) method and the Chi-Square (CS) method. The wrapper methods that have been considered include the Best First Search (BFS) method, the Linear Forward Selection (LFS) method and the Greedy Step Wise Search (GSS) method. A Decision Tree algorithm has been used as a classifier for this analysis and is implemented through the WEKA tool. An attribute significance analysis has been performed on the diabetes, breast cancer and heart disease datasets used in the study. It was observed that the CFS method outperformed other filter methods concerning the accuracy rate and execution time. The accuracy rate using the CFS method on the datasets for heart disease, diabetes, breast cancer was 93.8%, 89.5% and 96.8% respectively. Moreover, latency delays of 1.08 s, 1.02 s and 1.01 s were noted using the same method for the respective datasets. Among wrapper methods, BFS’ performance was impressive in comparison to other methods. Maximum accuracy of 94.7%, 95.8% and 96.8% were achieved on the datasets for heart disease, diabetes and breast cancer respectively. Latency delays of 1.42 s, 1.44 s and 132 s were recorded using the same method for the respective datasets. On the basis of the obtained result, a new hybrid Attribute Evaluator method has been proposed which effectively integrates enhanced K-Means clustering with the CFS filter method and the BFS wrapper method. Furthermore, the hybrid method was evaluated with an improved decision tree classifier. The improved decision tree classifier combined clustering with classification. It was validated on 14 different chronic disease datasets and its performance was recorded. A very optimal and consistent classification performance was observed. The mean values for accuracy, specificity, sensitivity and f-score metrics were 96.7%, 96.5%, 95.6% and 96.2% respectively.

Comparison of the number of nodes explored by cyclic best first search with depth contour and best first search

Search strategies are used in Branch-and-bound (B&B) algorithms to determine the order to process the nodes in a search tree. A search strategy that has seen frequent success in B&B algorithms is Best First Search (BFS). However, a new search strategy called Cyclic Best First Search with depth contour (CBFS-depth) has been shown to consistently process fewer nodes than BFS on several combinatorial optimization problems. In this paper, we study the properties of search trees under which CBFS-depth may outperform BFS in terms of the average number of nodes explored to prove optimality. To facilitate the study, a search tree model is proposed for minimization problems solved by B&B algorithms which is based on the distributions of nodes in the tree with different lower bounds. Using this model, we estimate the expected number of nodes expanded by CBFS-depth. Additionally, we apply B&B algorithms using both search strategies to trees that are randomly generated according to the proposed model in order to study how the average number of nodes expanded by each search strategy is affected by the search tree. Finally, both search strategies are tested on existing optimization problems to validate the observations from the randomly generated search trees.

An algorithm based on valuation forecasting for game tree search

This study proposes a novel best-first search algorithm, called Valuation Increment Search Algorithm (VIS), which applies the increment of exploratory value change to predict the valuation and guide the selection of the game tree. The proposed method can effectively use node valuation and game tree size information to solve the game tree by fewer exploration steps. The relevant assumptions and search principles of the proposed algorithm are detailed. Moreover, comparative experiments with Alpha–Beta Search (α–β search), Proof-Number Search (PNS) and Monte-Carlo Tree Search (MCTS) in the domain of Dou Dizhu has been performed to verify the performance. Result demonstrate that VIS is superior to α–β search, PNS and MCTS algorithm in solving the game tree and finding the best move for some game scenarios by consuming less time and few memory resources. The improvement effect of the average Increment, magnitude of Increment, and the number of visits to node on the original VIS was validated. Also, a new selection strategy is defined for the improved VIS algorithm combined with these factors. The experimental comparison results show that improved VIS has a better performance in solving game tree with less nodes generated and expanded than original VIS.

Two agents with GBFS algorithms working cooperatively to get a shortest path

This study is carried out in order to verify if the implementation of the concept of cooperative work among two agents, that use path planners A* to obtain the shortest path (previous work of the authors) is also valid when the cooperative strategy is applied using another path planner such as the so-called GBFS (Greedy Best First Search). In this sense, this paper shows a path planning strategy that combines the capabilities of two Agents each one with its own path planner GBFS (slightly different from each other) in order to obtain the shortest path. The comparisons between paths are made by analyzing the behavior and results obtained from the agents operating in different forms: (1) Working individually; (2) Working as a team (cooperating and exchanging information). The results show that in all analyzed situations are obtained shortest traveled distances when the path planners work as a cooperative team.

Relaxing Relationship Queries on Graph Data

In many domains we have witnessed the need to search a large entity-relation graph for direct and indirect relationships between a set of entities specified in a query. A search result, called a semantic association (SA), is typically a compact (e.g., diameter-constrained) connected subgraph containing all the query entities. For this problem of SA search, efficient algorithms exist but will return empty results if some query entities are distant in the graph. To reduce the occurrence of failing query and provide alternative results, we study the problem of query relaxation in the context of SA search. Simply relaxing the compactness constraint will sacrifice the compactness of an SA, and more importantly, may lead to performance issues and be impracticable. Instead, we focus on removing the smallest number of entities from the original failing query, to form a maximum successful sub-query which minimizes the loss of result quality caused by relaxation. We prove that verifying the success of a sub-query turns into finding an entity (called a certificate) that satisfies a distance-based condition about the query entities. To efficiently find a certificate of the success of a maximum sub-query, we propose a best-first search algorithm that leverages distance-based estimation to effectively prune the search space. We further improve its performance by adding two fine-grained heuristics: one based on degree and the other based on distance. Extensive experiments over popular RDF datasets demonstrate the efficiency of our algorithm, which is more scalable than baselines.

Machine Learning Based Statistical Analysis of Emotion Recognition using Facial Expression

Background: Humans can deliver many emotions during a conversation. Facial expressions show information about emotions.
 Objectives: This study proposed a Machine Learning (ML) approach based on a statistical analysis of emotion recognition using facial expression through a digital image.
 Methodology: A total of 600 digital image datasets divided into 6 classes (Anger, Happy, Fear, Surprise, Sad, and Normal) was collected from publicly available Taiwan Facial Expression Images Database. In the first step, all images are converted into a gray level format and 4 Regions of Interest (ROIs) are created on each image, so the total image dataset gets divided in 2400 (600 x 4) sub-images. In the second step, 3 types of statistical features named texture, histogram, and binary feature are extracted from each ROIs. The third step is a statistical feature optimization using the best-first search algorithm. Lastly, an optimized statistical feature dataset is deployed on various ML classifiers.
 Results: The analysis part was divided into two phases: firstly boosting algorithms-based ML classifiers (named as LogitBoost, AdaboostM1, and Stacking) which obtained 94.11%, 92.15%, and 89.21% accuracy, respectively. Secondly, decision tree algorithms named J48, Random Forest, and Random Committee were obtained with 97.05%, 93.14%, and 92.15% accuracy, respectively.
 Conclusion: It was observed that decision tree based J48 classifiers gave 97.05% classification accuracy.

Feature selection for improved classification accuracy targeting riverine sand mapping

Regular monitoring of riverine sand is crucial for its sustainable management. Towards readily generating spatial information regarding the extent of riverine sand, this study utilizes satellite-based remote sensing data as input features while applying Support Vector Machine (SVM) classifier to discern the inherent land cover classes. The input features comprised of spectral bands, derived spectral indices, derived textural features and ancillary information such as elevation, slope and aspect. The objective of this study was to identify a set of features that help improve the SVM classification accuracy with respect to the benchmark accuracy achieved by using only the spectral bands. Apart from testing a few commonly used logical combinations of features, the study focused on employing the Correlation-based Feature Selection (CFS) method along with the best-first search algorithm to generate a feature-set that comprised of most relevant and least correlated features. This feature-set was found to improve the SVM classification accuracy by 2.9% with respect to the benchmark value. Further, when tested with the limited number of training samples, similar results were achieved. This study proves that it is beneficial to utilize the CFS method for feature selection prior to SVM classification and recommends a set of 19 remote sensing derived features that are relevant towards riverine sand mapping.

The correlation structure of anomaly strategies

Abstract We consolidate a large number of mean-significant anomalies into cluster portfolios. More than a third of cluster portfolios remain significant under the Hou et al. (2020) five-factor model — the best performing among six benchmark models tested. A best-first search yields nine factors that subsume all cluster portfolios as well as all significant anomalies, demonstrating the feasibility of a parsimonious description of average realised returns. The expected growth factor (EG) and a cluster portfolio linked to accruals are prominent factors that improve pricing performance. The search-generated model produces a monthly maximum squared Sharpe ratio of 0.51, considerably higher than current benchmark models.

RIP based condition for support recovery with A* OMP in the presence of noise

A* orthogonal matching pursuit (A* OMP) aims at combination of best-first tree search with the OMP algorithm for the compressed sensing problem. In this study, the authors present a new analysis for the A* OMP algorithm using the restricted isometry property (RIP). The results show that if the sampling matrix A satisfies the RIP with δ K ⋆ < B / ( K + B ) ( K ⋆ = max { 2 K , K + B } ), then under some constraints on SNR, A* OMP accurately recovers the support of any K-sparse signal x from the samples y = A x + e , where B is the number of child paths for each candidate in the algorithm. In addition, the proposed condition is an optimal condition that guarantees the success of A* OMP in the noise-free case.

Finding the direct optimal RNA barrier energy and improving pathways with an arbitrary energy model

MotivationRNA folding kinetics plays an important role in the biological functions of RNA molecules. An important goal in the investigation of the kinetic behavior of RNAs is to find the folding pathway with the lowest energy barrier. For this purpose, most of the existing methods use heuristics because the number of possible pathways is huge even if only the shortest (direct) folding pathways are considered.ResultsIn this study, we propose a new method using a best-first search strategy to efficiently compute the exact solution of the minimum barrier energy of direct pathways. Using our method, we can find the exact direct pathways within a Hamming distance of 20, whereas the previous methods even miss the exact short pathways. Moreover, our method can be used to improve the pathways found by existing methods for exploring indirect pathways.Availability and implementationThe source code and datasets created and used in this research are available at https://github.com/eukaryo/czno. Supplementary information Supplementary data are available at Bioinformatics online.

Emotion Based Facial Expression Detection Using Machine Learning Approach

Humans have the capability to deliver many emotions during a conversation. Facial expressions show information about emotions. The major issue is to understand the facial expression during communication. Every face is an index of the mind. The objective of this study is to design a framework which has the ability to recognize human facial expression. Permanent and temporary facial expressions appear during conversation and detect using different face detection techniques. In this study, an emotion-based face identification system has been proposed by employing different machine learning approaches. Taiwanese Facial Expression Image Database (TFEID) has been used for three types of facial expression such as Angry, Happy and Sad. Each facial expression (Angry, Happy and Sad) contains 40 images and calculate total 120 (40 x 3) images dataset. For image pre-processing, Median filter has been employed on this dataset and converted color images to grayscale. Six non-overlapping regions of interest (ROIs) have been taken on every image and calculate 720 (120 x 6) ROIs on the overall dataset. Texture (T), Histogram (H) and Binary (B) features have been calculated on these three categories and extracted 43 features on each (ROIs) and calculated total 30960 (720 x 43) features vector pace on the deployed dataset. The Best First Search (BFS) algorithm has been implemented for feature optimization. The optimized dataset has been deployed to different machine learning classifiers namely Random Sub Space, Random Committee, Bagging, Random Forest, J48 and LMT. TreeRandom Forest has shown the best overall accuracy results among the deployed classifiers. The overall accuracy results of 95.277% has been observed by Tree Random Forest.

Emotion Based Facial Expression Detection Using Machine Learning Approach

Humans have the capability to deliver many emotions during a conversation. Facial expressions show information about emotions. The major issue is to understand the facial expression during communication. Every face is an index of the mind. The objective of this study is to design a framework which has the ability to recognize human facial expression. Permanent and temporary facial expressions appear during conversation and detect using different face detection techniques. In this study, an emotion-based face identification system has been proposed by employing different machine learning approaches. Taiwanese Facial Expression Image Database (TFEID) has been used for three types of facial expression such as Angry, Happy and Sad. Each facial expression (Angry, Happy and Sad) contains 40 images and calculate total 120 (40 x 3) images dataset. For image pre-processing, Median filter has been employed on this dataset and converted color images to grayscale. Six non-overlapping regions of interest (ROIs) have been taken on every image and calculate 720 (120 x 6) ROIs on the overall dataset. Texture (T), Histogram (H) and Binary (B) features have been calculated on these three categories and extracted 43 features on each (ROIs) and calculated total 30960 (720 x 43) features vector pace on the deployed dataset. The Best First Search (BFS) algorithm has been implemented for feature optimization. The optimized dataset has been deployed to different machine learning classifiers namely Random Sub Space, Random Committee, Bagging, Random Forest, J48 and LMT. TreeRandom Forest has shown the best overall accuracy results among the deployed classifiers. The overall accuracy results of 95.277% has been observed by Tree Random Forest.