Data Mining - Concepts and Techniques.

We explore the effect of dimensionality on the nearest neighbor problem. We show that under a broad set of conditions (much broader than independent and identically distributed dimensions), as dimensionality increases, the distance to the nearest data point approaches the distance to the farthest data point. To provide a practical perspective, we present empirical results on both real and synthetic data sets that demonstrate that this effect can occur for as few as 10-15 dimensions. These results should not be interpreted to mean that high-dimensional indexing is never meaningful; we illustrate this point by identifying some high-dimensional workloads for which this effect does not occur. However, our results do emphasize that the methodology used almost universally in the database literature to evaluate high-dimensional indexing techniques is flawed, and should be modified. In particular, most such techniques proposed in the literature are not evaluated versus simple linear scan, and are evaluated over workloads for which nearest neighbor is not meaningful. Often, even the reported experiments, when analyzed carefully, show that linear scan would outperform the techniques being proposed on the workloads studied in high (10-15) dimensionality!.

When is nearest neighbor meaningful

Search operations in databases require special support at the physical level. This is true for conventional databases as well as spatial databases, where typical search operations include the point query (find all objects that contain a given search point) and the region query (find all objects that overlap a given search region). More than ten years of spatial database research have resulted in a great variety of multidimensional access methods to support such operations. We give an overview of that work. After a brief survey of spatial data management in general, we first present the class of point access methods, which are used to search sets of points in two or more dimensions. The second part of the paper is devoted to spatial access methods to handle extended objects, such as rectangles or polyhedra. We conclude with a discussion of theoretical and experimental results concerning the relative performance of various approaches.

/pdf/multidimensional-access-methods-4jxurm5el0.pdf

Multidimensional access methods

本文对国际科学计量学杂志《Ｓｃｉｅｎｔｏｍｅｔｒｉｃｓ》１９７９－１９９１年的研究论文内容、栏目、作者及国别和编委及国别作了计量分析，揭示出科学计量学研究的重点、活动的中心及发展趋势，说明了学科带头人在发展科学计量学这门新兴学科中的作用。

国际刊物《Scientometrics》文献计量研究

During the last decade, multimedia databases have become increasingly important in many application areas such as medicine, CAD, geography, and molecular biology. An important research issue in the field of multimedia databases is the content-based retrieval of similar multimedia objects such as images, text, and videos. However, in contrast to searching data in a relational database, a content-based retrieval requires the search of similar objects as a basic functionality of the database system. Most of the approaches addressing similarity search use a so-called feature transformation that transforms important properties of the multimedia objects into high-dimensional points (feature vectors). Thus, the similarity search is transformed into a search of points in the feature space that are close to a given query point in the high-dimensional feature space. Query processing in high-dimensional spaces has therefore been a very active research area over the last few years. A number of new index structures and algorithms have been proposed. It has been shown that the new index structures considerably improve the performance in querying large multimedia databases. Based on recent tutorials [Berchtold and Keim 1998], in this survey we provide an overview of the current state of the art in querying multimedia databases, describing the index structures and algorithms for an efficient query processing in high-dimensional spaces. We identify the problems of processing queries in high-dimensional space, and we provide an overview of the proposed approaches to overcome these problems.

/pdf/searching-in-high-dimensional-spaces-index-structures-for-3dywul31y4.pdf

Searching in high-dimensional spaces: Index structures for improving the performance of multimedia databases

We propose a new multi-attribute index. Our approach combines the hB-tree, a multi-attribute index, and the $\Pi$-tree, an abstract index which offers efficient concurrency and recovery methods. We call the resulting method the hB $^\Pi$-tree. We describe several versions of the hB $^\Pi$-tree, each using a different node-splitting and index-term-posting algorithm. We also describe a new node deletion algorithm. We have implemented all the versions of the hB $^\Pi$-tree. Our performance results show that even the version that offers no performance guarantees, actually performs very well in terms of storage utilization, index size (fan-out), exact-match and range searching, under various data types and distributions. We have also shown that our index is fairly insensitive to increases in dimension. Thus, it is suitable for indexing high-dimensional applications. This property and the fact that all our versions of the hB $^\Pi$-tree can use the $\Pi$-tree concurrency and recovery algorithms make the hB $^\Pi$-tree a promising candidate for inclusion in a general-purpose DBMS.

/pdf/the-hb-pi-tree-a-multi-attribute-index-supporting-4gkxgztnqh.pdf

The hB $^\Pi$-tree: a multi-attribute index supporting concurrency, recovery and node consolidation

Vision Transformers (ViT) have shown rapid progress in computer vision tasks, achieving promising results on various benchmarks. However, due to the massive number of parameters and model design, \textit{e.g.}, attention mechanism, ViT-based models are generally times slower than lightweight convolutional networks. Therefore, the deployment of ViT for real-time applications is particularly challenging, especially on resource-constrained hardware such as mobile devices. Recent efforts try to reduce the computation complexity of ViT through network architecture search or hybrid design with MobileNet block, yet the inference speed is still unsatisfactory. This leads to an important question: can transformers run as fast as MobileNet while obtaining high performance? To answer this, we first revisit the network architecture and operators used in ViT-based models and identify inefficient designs. Then we introduce a dimension-consistent pure transformer (without MobileNet blocks) as a design paradigm. Finally, we perform latency-driven slimming to get a series of final models dubbed EfficientFormer. Extensive experiments show the superiority of EfficientFormer in performance and speed on mobile devices. Our fastest model, EfficientFormer-L1, achieves $79.2\%$ top-1 accuracy on ImageNet-1K with only $1.6$ ms inference latency on iPhone 12 (compiled with CoreML), which runs as fast as MobileNetV2$\times 1.4$ ($1.6$ ms, $74.7\%$ top-1), and our largest model, EfficientFormer-L7, obtains $83.3\%$ accuracy with only $7.0$ ms latency. Our work proves that properly designed transformers can reach extremely low latency on mobile devices while maintaining high performance.

EfficientFormer: Vision Transformers at MobileNet Speed

The dimensionality curse phenomenon states that in high dimensional spaces distances between nearest and farthest points from query points become almost equal. Therefore, nearest neighbor calculations cannot discriminate candidate points. Many indexing methods that try to cope with the dimensionality curse in high dimensional spaces have been proposed, but, usually these methods end up behaving like the sequential scan over the database in terms of accessed pages when queries like k-Nearest Neighbors are examined. In this paper, we experiment with state of the art multi-attribute indexing methods and try to investigate when these methods reach their limits, namely, at what dimensionality a kNN query requires visiting all the data pages. In our experiments we compare the Hybrid Tree, the R*-tree, and, the iDistance Method.

The Effects of Dimensionality Curse in High Dimensional kNN Search

The use of online videos is a common practice today amongst education professionals. Interactive features in videos are constantly evolving and a recent trend is the integration of interactive elements and web content into educational videos. The paper (a) provides a roadmap for using open source tools and open internet resources to develop a learning environment where video content is aggregated with interactive elements, educator content and content coming from the web, (b) describes how these open source tools are used for capturing and storing learner activity data, and, (c) presents findings obtained from analyzing learner activity data gathered in an educational setting during an academic year.

/pdf/using-open-source-technologies-and-open-internet-resources-1b130gfqtb.pdf

Using open source technologies and open internet resources for building an interactive video based learning environment that supports learning analytics

In Greece, the development of the teaching of Information Technology (IT) in schools has been greatly influenced by the rapid development of technology, making IT literacy a priority for all individuals. Consequently, the teaching of algorithms and programming, with the ulterior motive of teaching modeling as well as problem solving, has been greatly limited in Secondary Education. However, we strongly believe that algorithms and programming constitute an important intellectual tool and should be included in basic education. In any case, both the research literature findings, as well as, our own experience confirm the fact that novice programmers come up against many mental obstacles in their attempts to understand the functioning of programs or the construction of algorithms. In order to deal with these difficulties and successfully teach the elementary concepts of algorithms and programming, we have developed didactic scenarios, which are based on specially designed educational software. In conjunction with this, we are attempting to develop a program for the systematic training of those students who will become IT teachers in Secondary or Primary Education. The most significant findings of our research are summarized as follows: (a) The development of educational software and its experimental use in the teaching process allow us to formulate several general rules related to the specific didactic characteristics, which these environments should include. (b) Educational software is effective only when it is incorporated within the framework of the didactic scenarios that the teacher organizes and which are supported by the software. (c) Teachers do not spontaneously use educational software in the context of this rationale. Therefore, specific training is required so that they adopt and use such software in didactic scenarios.

Georgios Evangelidis

Papers

The hB $^\Pi$-tree: a multi-attribute index supporting concurrency, recovery and node consolidation

EfficientFormer: Vision Transformers at MobileNet Speed

The Effects of Dimensionality Curse in High Dimensional kNN Search

Using open source technologies and open internet resources for building an interactive video based learning environment that supports learning analytics

Introducing Secondary Education Students to Algorithms and Programming