Knowledge-Based Systems

It contains the results of the simulations to verify the results. Also, a GitHub repository is available to access to the source code of proposals.

It is generated as supplementary data in this work

This data set provides pricing details discussed in the article and explains the results shown in the research work.

The sentences from NYT corpus of the years 2005 and 2006 are aligned by entities in Freebase for training, while that of the year 2007 are aligned for testing. The dataset has 52 common relations and a special relation NA that there is no relation between entity pair e1 and e2. Overall, there are 522,611 sentences, 281,270 entity pairs and 18,252 relational facts in training data, and 172,448 sentences, 96,678 entity pairs and 1,950 relational facts in test data, respectively.

This dataset contains the muti-turn user-machine utterances with IOB representation for slot values and intent annotation.

The datasets include the adjusted daily and weekly closing prices of assets enlisted in DOW30, NASDAQ100, and S&P500 stock indices.

This datasets contains the answers to the NASA Task Load index questionnaire (columns E-Z) from a set of university students, as well as the ID of the administered questionnaire (column A), a description of the taught topic (column B), the instructional delivery method used (column C), and the self-reported uni-dimensional perception of mental workload reported by students (column D).

The dataset can be used as the benchmark dataset for the researcher of machine learning or related areas.

1. Training Datasets: 1.1. Columns details: columns represent the extracted features from the content of source profile (Twitter) and the target profiles (Facebook). Last column shows the match and no-match condition. Total number of columns in this dataset are 31. 1.2. Rows details: all the rows in this dataset represent the source and target profiles pairs for match and no-match. Total number of rows in this dataset are 31882. 2. Test Datasets: 2.1. Columns details: columns represent the extracted features from the content of source profile (Twitter)and the target profiles (Facebook). Last column shows the match and no-match condition. Total number of columns in this dataset are 31. 2.2. Rows details: all the rows in this dataset represent the source and target profiles pairs for match and no-match. Total number of rows in this dataset are 17392.

These datasets were used for the experimental evaluation of the Hierarchical Distributed Stream Miner (HDSM) as described in the accompanying article. These datasets may be useful for evaluating other distributed data stream mining methods.