Large-scale Data Exploration Using Explanatory Regression Functions

Abstract

Analysts wishing to explore multivariate data spaces, typically issue queries involving selection operators, i.e., range or equality predicates, which define data subspaces of potential interest. Then, they use aggregation functions, the results of which determine a subspace's interestingness for further exploration and deeper analysis. However, Aggregate Query (AQ) results are scalars and convey limited information and explainability about the queried subspaces for enhanced exploratory analysis. Analysts have no way of identifying how these results are derived or how they change w.r.t query (input) parameter values. We address this shortcoming by aiding analysts to explore and understand data subspaces by contributing a novel explanation mechanism based on Machine Learning. We explain AQ results using functions obtained by a threefold joint optimization problem which assume the form of explainable piecewise-linear regression functions. A key feature of the proposed solution is that the explanation functions are estimated using past executed queries. These queries provide a coarse grained overview of the underlying aggregate function (generating the AQ results) to be learned. Explanations for future, previously unseen AQs can be computed without accessing the underlying data and can be used to further explore the queried data subspaces, without issuing more queries to the backend analytics engine. We evaluate the explanation accuracy and efficiency through theoretically grounded metrics over real-world and synthetic datasets and query workloads. CCS Concepts: • Mathematics of computing → Exploratory data analysis; • Information systems → Data analytics; • Computing methodologies → Supervised learning by regression.

Query-driven Edge Node Selection in Distributed Learning Environments

“…Hence, a node selection mechanism is required to determine which node could improve the model performance and which can lead to a model that can forget what it has learned. Unfortunately, upon any incoming analytics query [13], we do not have full access to data of all participants; thus, we cannot extract the data space and patterns. In order to extract this knowledge per incoming query, it is deemed appropriate to define a pre-test mechanism to check if participants have a similar or different data patterns and relevant to the current query or not.…”

CFLCA: High Performance based Heart disease Prediction System using Fuzzy Learning with Neural Networks

“…For a particular data mining application, the importance of the prediction and description may differ some times. The goals of description and prediction are accomplished by using variety of mining methods of data [7] [8].…”

Max-Utility Based Arm Selection Strategy For Sequential Query Recommendations

“…Typical use case scenarios include: (i) data scientists analyzing a large volume of data to obtain in-depth knowledge about the data for follow-up tasks like, data trend explanation (Savva et al, 2020), report summarization (Marcel and Negre, 2011) and (ii) users gathering information by discovering scholarly articles to conduct literature review using online services (Krause and Guestrin, 2011). These tasks can be conceptualized as the system recommending queries and the user either accepting or rejecting these recommendations, thus forming a closed loop interactive environment.…”