The geological big data is a kind of spatio-temporal big data whose characteristics has both similarities and marked differences with the general big data. Due to long-term evolution of geological objects with huge space and influence of various geological process, its development undergoes a complicated process. Geological bodies are buried in deep underground so that geological data has the feature of incomplete information of parameter, structure, relationship and evolution, which has high computation complexity with high dimension and significant uncertainty. In order to address those challenges, the theories, methods and techniques of big data should be introduced to integrate and utilize geological science big data. It is necessary to set up unified spatial reference system for geological spatio-temporal big data to realize data consistency processing and fusion with regards to spatial standard benchmark, tense, scale and semantics. Various kinds of integrated storage and management methods for static geological exploration data and dynamic distributed geological observation data will be developed. On the one hand, current big data techniques can be adapted, such as realtime data storage and management, big data storage and management architecture, and analysis and processing architecture. On the other hand, specialized big data techniques based on actual situation should be developed, such as distributed parallel spatio-temporal indexing for geological data, and pre-dispatch method considering spatio- temporal relationship and geological semantics. To judge a traditional algorithm whether is suitable for being transformed into big data environment, main criteria include algorithm task decomposability, data decomposability and data flow segmented relevance. Hence, the transformation of traditional geological data processing methods will be from these three aspects to excavate the task parallelism, geometry decomposability and reduce the data flow segmented relevance as much as possible to adapt to the distributed computing and high performance computing of big data environment. The big data technique, which may be a breakthrough in geological science, can mine knowledge from massive spatio-temporal and text data directly in spite of the limits in geological sampling methods such as the randomness and sparsity of sample space, the judge relied on inadequate observational data and fixed mode, and traditional data analysis methods. A series of mathematical geology methods and spatial data mining methods which are frequently used nowadays can be used in geological spatio-temporal big data mining after being remolded. The key science and technology issues referred in geological science development in big data era can be listed as follows—the integrated storage management and handling of structured, semi-structured and unstructured data, big and small data, hybrid and precision data, model and data, static exploration model and dynamic monitoring model, the combination of data mining and data analysis, the unification of correlation and causality, deep mining and visualization of geological science big data, and so on. Taking metal mineral metallogenic prediction as an example, geological big data utilization and analysis are studied. In terms of a global view, Glass Earth is the underground extention of Digital Earth. Glass Earth is a kind of 3D visualized virtual shallow crust, which is an aggregation of geological and geographical information and stored on the computer network to be accessed by multi-user and provide analysis service and decision making to the research and applications of geology, resources and environment. The Glass Earth is the effective vehicle of geological science big data and the construction of the Glass Earth is one of the best ways to solve all the aforementioned scientific and technical problems.

A large amount of time series data is being generated every day in a wide range of sensor application domains. The symbolic aggregate approximation (SAX) is a well-known time series representation method, which has a lower bound to Euclidean distance and may discretize continuous time series. SAX has been widely used for applications in various domains, such as mobile data management, financial investment, and shape discovery. However, the SAX representation has a limitation: Symbols are mapped from the average values of segments, but SAX does not consider the boundary distance in the segments. Different segments with similar average values may be mapped to the same symbols, and the SAX distance between them is 0. In this paper, we propose a novel representation named SAX-BD (boundary distance) by integrating the SAX distance with a weighted boundary distance. The experimental results show that SAX-BD significantly outperforms the SAX representation, ESAX representation, and SAX-TD representation.

Timelines have been used for centuries and have become more and more widely used with the development of social media in recent years. Every day, various smart phones and other instruments on the internet of things generate massive data related to time. Most of these data can be managed in the way of timelines. However, it is still a challenge to effectively and efficiently store, query, and process big timeline data, especially the instant recommendation based on timeline similarities. Most existing studies have focused on indexing spatial and interval datasets rather than the timeline dataset. In addition, many of them are designed for a centralized system. A timeline index structure adapting to parallel and distributed computation framework is in urgent need. In this research, we have defined the timeline similarity query and developed a novel timeline index in the distributed system, called the Distributed Triangle Increment Tree (DTI-Tree), to support the similarity query. The DTI-Tree consists of one T-Tree and one or more TI-Trees based on a triangle increment partition strategy with the Apache Spark. Furthermore, we have provided an open source timeline benchmark data generator, named TimelineGenerator, to generate various timeline test datasets for different conditions. The experiments for DTI-Tree’s construction, insertion, deletion, and similarity queries have been executed on a cluster with two benchmark datasets that are generated by TimelineGenerator. The experimental results show that the DTI-tree provides an effective and efficient distributed index solution to big timeline data.

Time series data are widely found in finance, health, environmental, social, mobile and other fields. A large amount of time series data has been produced due to the general use of smartphones, various sensors, RFID and other internet devices. How a time series is represented is key to the efficient and effective storage and management of time series data, as well as being very important to time series classification. Two new time series representation methods, Hexadecimal Aggregate approXimation (HAX) and Point Aggregate approXimation (PAX), are proposed in this paper. The two methods represent each segment of a time series as a transformable interval object (TIO). Then, each TIO is mapped to a spatial point located on a two-dimensional plane. Finally, the HAX maps each point to a hexadecimal digit so that a time series is converted into a hex string. The experimental results show that HAX has higher classification accuracy than Symbolic Aggregate approXimation (SAX) but a lower one than some SAX variants (SAX-TD, SAX-BD). The HAX has the same space cost as SAX but is lower than these variants. The PAX has higher classification accuracy than HAX and is extremely close to the Euclidean distance (ED) measurement; however, the space cost of PAX is generally much lower than the space cost of ED. HAX and PAX are general representation methods that can also support geoscience time series clustering, indexing and query except for classification.

Three-dimensional voxel models are widely applied in various fields such as 3D imaging, industrial design, and medical imaging. The advancement of 3D modeling techniques and measurement devices has made the generation of three-dimensional models more convenient. The exponential increase in the number of 3D models presents a significant challenge for model retrieval. Currently, these models are numerous and typically represented as point clouds or meshes, resulting in sparse data and high feature dimensions within the retrieval database. Traditional methods for 3D model retrieval suffer from high computational complexity and slow retrieval speeds. To address this issue, this paper combines spatial-filling curves with octree structures and proposes a novel approach for representing three-dimensional voxel model sequence data features, along with a similarity measurement method based on symbolic operators. This approach enables efficient similarity calculations and rapid dimensionality reduction for the three-dimensional model database, facilitating efficient similarity calculations and expedited retrieval.