Top-k High Utility Itemset Mining Research Articles

Efficient mining of top-k high utility itemsets through genetic algorithms

Mining high utility itemsets is an emerging and very active research area in data mining. The goal is to mine all itemsets with a utility value, in terms of importance to the user, no less than a predefined threshold value. Setting an appropriate threshold value is not trivial, requiring not only multiple trials but also the know-how in the application field. The advantage of algorithms for mining top-k high utility itemsets is they do not require such a utility threshold, but they suffer from very long runtimes and large memory requirements when large input data is considered. We propose a new genetic algorithm for mining top-k high utility itemsets, named TKHUIM-GA (Top-K High Utility Itemset Mining through Genetic Algorithms). It guides the search process by considering the utility of each item to produce initial solutions and to combine solutions accordingly, reducing the runtime and memory consumption as a result. A highly efficient data representation is utilized to reduce memory usage and runtime. A key advantage of TKHUIM-GA is that it works on positive, negative, integer and real unit utility values unlike existing approaches. Experiments on popular benchmark datasets demonstrate the high performance of the proposal regarding the state-of-the-art algorithms.

Efficient top-k high utility itemset mining on massive data

In practical applications, top-k high utility itemset mining (top-k HUIM) is an interesting operation to find the k itemsets with the highest utilities. It is analyzed that, the existing algorithms only can deal with the small and medium-sized data, and their performance degrades significantly on massive data. This paper presents a novel top-k HUIM algorithm PTM which can mine top-k high utility itemsets on massive data efficiently. PTM maintains the transaction database as a set of prefix-based partitions, each of which can fit in the memory and keeps the transactions with the same prefix item. The utility of an itemset can be computed by the transactions in one particular partition. PTM processes the prefix-based partitions in the order of average transaction utility to raise utility threshold faster. By the concise assistant data structures, PTM can skip majority of the partitions directly. For the partitions to be processed, PTM utilizes depth-first search on the set enumeration tree of the promising items to find the required results. The full-suffix-utility-based subtree pruning rule is devised to reduce the exploration space of set enumeration tree effectively. The extensive experimental results show that PTM can discover the top-k high utility itemsets on massive data efficiently.

Mining of top-k high utility itemsets with negative utility

High utility itemset mining (HUIM) with negative utility is an emerging data mining task. However, the setting of the minimum utility threshold is always a challenge when mining high utility itemsets (HUIs) with negative items. Although the top-k HUIM method is very common, this method can only mine itemsets with positive items, and the problem of missing itemsets occurs when mining itemsets with negative items. To solve this problem, we first propose an effective algorithm called THN (Top-k High Utility Itemset Mining with Negative Utility). It proposes a strategy for automatically increasing the minimum utility threshold. In order to solve the problem of multiple scans of the database, it uses transaction merging and dataset projection technology. It uses a redefined sub-tree utility value and a redefined local utility value to prune the search space. Experimental results on real datasets show that THN is efficient in terms of runtime and memory usage, and has excellent scalability. Moreover, experiments show that THN performs particularly well on dense datasets.

Top-k High Utility Itemset Mining Framework by using TKU Along with TKO Algorithms

High utility itemsets (HUIs) mining be developing point within in sequence mining, which alludes toward discovering the entire itemsets contains utility assembling consumer verified slightest utility farthest point minutil. In any case, setting minutil fittingly is a problematic issue for customers. When in doubt, discovering reasonable smallest amount utility farthest point through testing be repetitive methodology intended for customers. On off chance to minutil be place excessively small, such countless HUIs determination exist made, which may make mining technique survive uncommonly incompetent. Additional supply taking place, condition minutil be place unnecessarily, we do not find any HUI. Within term document, we deal with the exceeding concerns through planning an additional composition in favor of top-k high utility itemset mining, where k be an ideal number of HUIs toward mined. Two arranges about viable counts named TKU (mining Top-K Utility itemsets) plus TKO (mining Top-K utility itemsets within one phase) be planned in favor of mining that itemsets not including require toward position minutil. We provide assistant relationship about two estimations among trades on top of central focuses along with obstacles. Precise appraisals lying on together authentic as well as made datasets shows introduction about proposed computations be close to the perfect case about top tier utility mining estimations.

Mining top-k high utility itemsets with effective threshold raising strategies

Top-K High Utility Itemset (HUI) mining problem offers greater flexibility to a decision maker in specifying her/his notion of item utility and the desired number of patterns. It obviates the need for a decision maker to determine an appropriate minimum utility threshold value using a trial-and-error process. The top-k HUI mining problem, however, is more challenging and requires use of effective threshold raising strategies. Several threshold raising strategies have been proposed in the literature to improve the overall efficiency of mining top-k HUIs. This paper advances the state-of-the-art and presents a new Top-K HUI method (THUI). A novel Leaf Itemset Utility (LIU) structure and a threshold raising strategy is proposed to significantly improve the efficiency of mining top-k HUIs. A new utility lower bound estimation method is also introduced to quickly raise the minimum utility threshold value. The proposed THUI method is experimentally evaluated on several benchmark datasets and compared against two state-of-the-art methods. Our experimental results reveal that the proposed THUI method offers one to three orders of magnitude runtime performance improvement over other related methods in the literature, especially on large, dense and long average transaction length datasets. In addition, the memory requirements of the proposed method are found to be lower.

Mining top-k high-utility itemsets from a data stream under sliding window model

High-utility itemset mining has gained significant attention in the past few years. It aims to find sets of items i.e. itemsets from a database with utility no less than a user defined threshold. The notion of utility provides more flexibility to an analyst to mine relevant itemsets. Nowadays, a continuous and unbounded stream of data is generated from web-clicks, transaction flow from retail stores, sensor networks, etc. Mining high-utility itemsets from a data stream is a challenging task as the incoming stream of data has to be processed on the fly with time and storage memory constraints. The number of high-utility itemsets depends on the user-defined threshold. A large number of itemsets can be generated at very low threshold values and vice versa. It can be a tedious task to set a threshold value to get a reasonable number of itemsets. Top-k high-utility itemset mining was coined to address this issue. k is the number of high-utility itemsets in the result set as defined by the user. In this paper, we propose a data structure and an efficient algorithm for mining top-k high-utility itemsets from a data stream. The algorithm has a single phase that does not generate any candidates, unlike many algorithms that work in two phases, i.e., candidate generation followed by candidates verification. We conduct extensive experiments on several real and synthetic datasets. Experimental results demonstrate that our proposed algorithm performs 20 to 80 times better on sparse datasets and 300 to 700 times on dense datasets than the state-of-the-art algorithm in terms of computation time. Furthermore, our proposed algorithm requires less memory compared to the state-of-the-art algorithm.

An efficient algorithm for mining the top-k high utility itemsets, using novel threshold raising and pruning strategies

Top-k high utility itemset mining is the process of discovering the k itemsets having the highest utilities in a transactional database. In recent years, several algorithms have been proposed for this task. However, it remains very expensive both in terms of runtime and memory consumption. The reason is that current algorithms often generate a huge amount of candidate itemsets and are unable to prune the search space effectively. In this paper, we address this issue by proposing a novel algorithm named kHMC to discover the top-k high utility itemsets more efficiently. Unlike several algorithms for top-k high utility itemset mining, kHMC discovers high utility itemsets using a single phase. Furthermore, it employs three strategies named RIU, CUD, and COV to raise its internal minimum utility threshold effectively, and thus reduce the search space. The COV strategy introduces a novel concept of coverage. The concept of coverage can be employed to prune the search space in high utility itemset mining, or to raise the threshold in top-k high utility itemset mining, as proposed in this paper. Furthermore, kHMC relies on a novel co-occurrence pruning technique named EUCPT to avoid performing costly join operations for calculating the utilities of itemsets. Moreover, a novel pruning strategy named TEP is proposed for reducing the search space. To evaluate the performance of the proposed algorithm, extensive experiments have been conducted on six datasets having various characteristics. Results show that the proposed algorithm outperforms the state-of-the-art TKO and REPT algorithms for top-k high utility itemset mining both in terms of memory consumption and runtime.

An Algorithm of Top-k High Utility Itemsets Mining over Data Stream

Existing top-k high utility itemset (HUI) mining algorithms generate candidate itemsets in the mining process; their time & space performance might be severely affected when the dataset is large or contains many long transactions; and when applied to data streams, the performance of corresponding mining algorithm is especially crucial. To address this issue, propose a sliding window based top-k HUIs mining algorithm TOPK-SW; it first stores each batch data of current window as well as the items' utility information to a tree called HUI-Tree, which ensures effective retrieval of utility values without re-scan the dataset, so as to efficiently improve the mining performance. TOPK-SW was tested on 4 classical datasets; results show that TOPK-SW outperforms existing algorithms significantly in both time and space efficiency, especially the time performance improves over 1 order of magnitude.