Machine learning has become the cornerstone of large data

Machine learning is almost ubiquitous, and even if we don't call them, they often appear in large data applications. I used to describe some typical big data use cases in my blog. In other words, these applications can provide the best results in "extreme situations". At the end, I also mentioned the combination of byte-level data capacity, real-time data speed, and/or diversity of multiple structured data.

I also listed a list of applications that deliberately avoided "machine learning analysis" during the collection process. The main reason is that although machine learning is a common tool in these use cases, it is not a use case. In other words, they are not a special application domain formed by their own power. For the same reason, I don't list schema design, meta data management, or data consolidation like big data use cases. But like machine learning, they have made their contribution to the value of large data analysis applications.

The contribution of machine learning to the investment return of large data application is mainly embodied in two aspects: one is to promote the productivity of the data scientists, the other is to find some neglected schemes, some of which have even been overlooked by the best data scientists. These values derive from the core functions of machine learning: the ability to learn the latest data without human intervention and explicit procedures. The solution allows data scientists to create a model based on a typical dataset, and then automate the generalization and learning of these examples and new data sources using algorithms.

In many cases, machine learning is the best investment return for large data innovations. Investing in machine learning can deepen any large data case that is customized to the enterprise. This is because machine learning algorithms are becoming more efficient in capacity, speed, and type (that is, 3 v characteristics of large data). As Mark Van Rijmenam in a recent article on machine learning: "The more data is processed, the better the algorithm will be." He believes that many machine learning applications, including voice and facial recognition, click Stream Processing, search engine optimization, and recommendation engines, may be described as sense-making Analytics.

The tacit analysis method needs to continuously monitor the user semantic way, content and importance from the data stream. In order to support the automation of tacit knowledge, machine learning algorithms must deal with some extremely complex things frequently. This includes semantic classifications that are hidden in the composition object or environment, which requires real-time collection of the whole meaning through a variety of different data streams. These data streams must include elements such as data, video, images, voice, expressions, actions, geographic information, and browser clicks. The automatic extraction of the meaning from these data streams through machine learning may be mixed with cognitive, emotional, sensory and will characteristics.

In order to find clues in these materials, "deep learning" (deep learning) has become an important tool in the machine learning instruction system of large data scientists. As Van Rijmenam says, deep learning using neural networks can help extract perceptual capabilities from these streams of data that may involve hierarchical arrangements for semantic relationships between objects. "Deep learning can break the gap between components that have different characteristics in the data and use these features to find different combinations of features to figure out what they see or do." Van Rijmenam said.

Clearly, machines learn a fundamental tool for creating environments that can perceive and process dynamic distributed scenarios. The ability of humans to detect and respond to real-time threats and terrorist activities, natural disasters, hurricanes and other threats depends on the automatic screening, classification and association of Information in mass data. Without this ability, humans are at risk of drowning in large data oceans.

36 Large Data Knowledge Map: About machine learning

Machine learning (Machine Learning, ML) is a multidisciplinary interdisciplinary, involving a number of disciplines such as probability theory, statistics, approximation theory, convex analysis and algorithmic complexity theory. Specialized in studying how computers simulate or implement human learning behavior in order to acquire new knowledge or skills, and to rearrange existing knowledge structures to continuously improve their performance.

It is the core of artificial intelligence, is to make the computer has the basic way of intelligence, its application in all fields of artificial intelligence, it mainly uses induction, synthesis rather than deduction.