Download JAR files tagged by distance with all dependencies

This module provides two packages, 'trees' and 'forests', which provide algorithms to compute edit distances on trees and forests (that is, unordered or ordered lists of trees) respectively. The edit distance is computed according to the tree edit distance algorithm of Zhang and Shasha (1989). The basic tree data structure is defined by the Tree interface in the trees module. Please refer to the javadoc for more detailed information.

This package contains several example programs of the usage of the JavaPermutationTools (JPT) library. JPT is a Java library that enables representing and generating permutations and sequences, as well as performing computation on permutations and sequences. It includes implementations of a variety of permutation distance metrics as well as distance metrics on sequences (i.e., Strings, arrays, and other ordered data types). In addition to programs demonstrating the usage of the JPT library, the jpt-examples package also contains programs for replicating the experiments from a few published papers that utilized the library or implementations on which the library is based. JPT's source code is maintained on GitHub, and the prebuilt jars of the library can be imported from Maven Central using maven or other build tools. The purpose of the package jpt-examples is to demonstrate usage of the major functionality of the JPT library. You can find out more about the JPT library itself from its website: https://jpt.cicirello.org/.

This module is a custom implementation of the Large Margin Nearest Neighbor classification scheme of Weinberger, Saul, et al. (2009). It contains an implementation of the k-nearest neighbor and LMNN classifier as well as (most importantly) gradient calculation schemes on the LMNN cost function given a sequential data set and a user-choice of alignment algorithm. This enables users to learn parameters of the alignment distance in question using a gradient descent on the LMNN cost function. More information on this approach can be found in the Masters Thesis "Adaptive Affine Sequence Alignment Using Algebraic Dynamic Programming"

FunctionExecutors 1. GeoWithinFunctionExecutor Input : (longitude double, latitude double, geoJSONGeometryFence string) OR (geoJSONGeometry string, geoJSONGeometryFence string) Output : true if (longitude, latitude) or geoJSONGeometry is within the geoJSONGeometryFence 2. GeoIntersectsFunctionExecutor Input : (longitude double, latitude double, geoJSONGeometryFence string) OR (geoJSONGeometry string, geoJSONGeometryFence string) Output : true if (longitude, latitude) or geoJSONGeometry intersects the geoJSONGeometryFence 3. GeoWithinDistanceFunctionExecutor Input : (longitude double, latitude double, geoJSONGeometryFence string, distance double) OR (geoJSONGeometry string, geoJSONGeometryFence string, distance double) Output : true if (longitude, latitude) or geoJSONGeometry is within distance of the geoJSONGeometryFence StreamProcessors 1. GeoCrossesStreamProcessor Input : (id string, longitude double, latitude double, geoJSONGeometryFence string) OR (id string, geoJSONGeometry string, geoJSONGeometryFence string) Output : an event with `crosses` additional attribute set to true when the object ((longitude, latitude) or geoJSONGeometry) crosses into geoJSONGeometryFence and an event with `crosses` additional attribute set to false when the object crosses out of the geoJSONGeometryFence 2. GeoStationaryStreamProcessor Input : (id string, longitude double, latitude double, geoJSONGeometryFence string, radius double) OR (id string, geoJSONGeometry string, geoJSONGeometryFence string, radius double) Output : when the object ((longitude, latitude) or geoJSONGeometry) starts being stationary within the radius an event with `stationary` additional attribute set to true. When the object starts to move out of the radius an event with `stationary` additional attribute set to false. 3. GeoProximityStreamProcessor Input : (id string, longitude double, latitude double, geoJSONGeometryFence string, radius double) OR (id string, geoJSONGeometry string, geoJSONGeometryFence string, radius double) Output : when two objects ((longitude, latitude) or geoJSONGeometry) starts being in close proximity within the radius an event with `inCloseProximity` additional attribute set to true. When the object starts to move out of the radius an event with `inCloseProximity` additional. attribute set to false. On each event, additional attributes `proximityWith` gives the id of the object that this object is in close proximity and `proximityId` is an id unique to the pair of objects

A filter that applies the LOF (Local Outlier Factor) algorithm to compute an outlier score for each instance in the data. Can use multiple cores/cpus to speed up the LOF computation for large datasets. Nearest neighbor search methods and distance functions are pluggable. For more information, see: Markus M. Breunig, Hans-Peter Kriegel, Raymond T. Ng, Jorg Sander (2000). LOF: Identifying Density-Based Local Outliers. ACM SIGMOD Record. 29(2):93-104.

YET ANOTHER IMPLEMENTATION OF LEVENSHTEIN DISTANCE GenericLevenshtein is an implementation of Minimum Edit Distance, also called Levenshtein Distance, written by Ramon Ziai and Niels Ott. This algorithm is very popular and it is often used to compute the similarity of strings. The difference in the presented implementation is that it can operate on sequences of any Java object implementing equals(Object). So no matter if you want to compare genome sequences or sequences of numbers, or just strings, here you go! Furthermore, the costs of the replace, insert, and delete operations can be customized by implementing the simple WeightCalculator<T> interface. In that case it is not a requirement to rely on equals(Object) as your implementation can do whatever you like it to do in oder to compare objects.

Cluster data using the sequential information bottleneck algorithm. Note: only hard clustering scheme is supported. sIB assign for each instance the cluster that have the minimum cost/distance to the instance. The trade-off beta is set to infinite so 1/beta is zero. For more information, see: Noam Slonim, Nir Friedman, Naftali Tishby: Unsupervised document classification using sequential information maximization. In: Proceedings of the 25th International ACM SIGIR Conference on Research and Development in Information Retrieval, 129-136, 2002.

This is an implementation of the Neural Gas algorithm on distance data (Relational Neural Gas) for unsupervised clustering. We recommend that you use the functions provided by the RelationalNeuralGas class for your purposes. All other classes and functions are utilities which are used by this central class. In particular, you can use RelationalNeuralGas.train() to obtain a RNGModel (i.e. a clustering of your data), and subsequently you can use RelationalNeuralGas.getAssignments() to obtain the resulting cluster assignments, and RelationalNeuralGas.classify() to cluster new points which are not part of the training data set. The underlying scientific work is summarized nicely in the dissertation "Topographic Mapping of Dissimilarity Datasets" by Alexander Hasenfuss (2009). The basic properties of an Relational Neural Gas algorithm are the following: 1.) It is relational: The data is represented only in terms of a pairwise distance matrix. 2.) It is a clustering method: The algorithm provides a clustering model, that is: After calculation, each data point should be assigned to a cluster (for this package here we only consider hard clustering, that is: each data point is assigned to exactly one cluster). 3.) It is a vector quantization method: Each cluster corresponds to a prototype, which is in the center of the cluster and data points are assigned to the cluster if and only if they are closest to this particular prototype. 4.) It is rank-based: The updates of the prototypes depend only on the distance ranking, not on the absolute value of the distances.

Includes 2 main parts: - Path finding through 2D polygons using the A star algorithm and navigation-mesh generation Field of vision / shadows / line of sight / lighting. The basic polygon and point classes are the KPolygon and KPoint. KPolygon contains a list of KPoints for vertices as well as a center (centroid), area, and radius (circular bound or distance from center to furthest point). KPolygon was born out of the need for a more game-oriented and flexible polygon class than the Path2D class in the standard Java library. KPolygon implements java.awt.geom.Shape so it can be easily drawn and filled by Java2D's Graphics2D object. - This API provides path-finding and field-of-vision. For other complex geometric operations such as buffering (fattening and shrinking) and constructive area geometry (intersections and unions) it is recommended to use the excellent Java Topology Suite (JTS). The standard Java2D library also provides the Area class which can be used for some constructive area geometry operations. Note that there is a utility class PolygonConverter that can quickly convert KPolygons to JTS polygons and vice versa.