Shulou(Shulou.com)06/02 Report--

Activity description: the kicking tribe is a service community that provides core developers with in-depth technical exchanges, development needs and resource sharing. Based on this community, we have invited technology leaders in the industry to make one-on-one breakthroughs in the development needs and remove the stumbling blocks in the development process. Solve development problems for developers with the most professional and efficient answers.

Topic keywords: big data Medical Mobile Android Internet of things machine learning

Tribal lineup: Luo Qibin, Chiyunod CEO, chief editor of Genespace

Object oriented: front-end development engineer, Android development engineer, product manager

How to participate: join 51CTO developer QQ exchange group 370892523, if you have any technical questions, ask questions in the group, or send it to the group owner and junior official.

Event details:

Shanghai-Xu Liangliang-products: I would like to pay attention to some knowledge related to data analysis, especially some applications and practices in the medical field.

Luo Qibin: the difficulty in combining the medical field with big data is the quality of its data. Because most of the algorithms and its technology are actually traditional sunshine intelligence or machine learning algorithms, there is no big improvement in this algorithm, but at the high-quality data level of data acquisition, it is now a global bottleneck. All the medical data you see with breakthroughs are basically based on high-quality data when applied to these good artificial intelligence algorithms. If anyone has a source of high-quality data, it is actually a great advantage of scarce resources, so many people are now starting to attack the construction and investment of a biological resource and biological sample storehouse. that is, once who occupies the entrance to this high-quality data, then all the technical personnel are actually gathered on this entrance. So this is a specific effect of the platform.

Shanghai-Xu Liangliang-products: I would like to know what are the current enterprise applications of big data or data analysis?

Luo Qibin: like what we are doing now, it is an enterprise-level data analysis product. Open accurate health management, provide one-stop bioinformatics big data platform, and provide the most comprehensive bioinformatics analysis applications for users in scientific research and medical fields.

Shanghai-Xu Liangliang-products: how to obtain the relevant requirements of enterprise data analysis and then convert them into related product functions? For example, you can now access the city's personal health file data and chronic disease data, but it is still just a business system. In the later stage, there is no train of thought to do the analysis based on these data. Mainly consider how to serve the residents, or the relevant functional departments of the Health and Family Planning Commission, hoping that professionals can give some ideas and my later learning direction.

Luo Qibin: like this kind of construction problem that belongs to a personal health file, it is a bit like an intelligent community. This construction process can not be done in one step, but a long-term engineering problem. Well, many of the previous stages are in fact a basic architectural construction, for example, the hardware facilities at the bottom, that is, from the IaaS layer to the middle PaaS layer, to the upper SaaS layer, in fact, the construction of each layer has experienced such a stage in the data application of this health file. When can we use genetic data or use a medical data? this is done in stages. And each stage of its data acquisition, there is a process.

Chongqing-Xiao Wang-soft Research: the Internet of things and the development status of big data?

51CTO: we have shared, you can refer to: the Internet of things and the opportunities and challenges faced by big data era

Chengdu-Fomkys- backend: is there any information about machine learning? Mainly want to know more about machine learning related knowledge.

51CTO: there are many machine learning materials on our platform, such as:

13 traps to pay attention to in machine learning! The 5 most common mistakes made by rookies in machine learning and how to avoid them

A Summary of gradient Descent Optimization algorithm, one of the most commonly used Optimization in Machine Learning

Chongqing-package-the method of keeping Android:Android process alive. Well, the company requires that the APP we do can always run in the background without being killed like QQ, that is, the Android process can be kept alive. At present, we use to start a foreground Service process, which will generate a Notification in the notification bar of the system, which is used to let users know that such an app is running, but in the case of insufficient system memory or after a period of time. The system begins to kill some processes according to its own set of process recovery mechanism. Then our program is recycled.

51CTO: there is a notification bar that basically will not be recycled. For example, if you play music, it will never stop.

Beijing-Zhang Xin-Front end: who knows how to import react-native android projects with android studio?

The android project generated by 51CTO:react native can be opened directly with android studio.

Beijing-Song Junjun-Android developer: Android performance is a big problem with Android's ANR and memory leak and memory fluctuation solutions.

51CTO: these problems can be avoided by following the development principles of Google. It doesn't take long to occupy the main thread to avoid ANR. Memory leaks can be avoided by not strongly referencing the context (except for system bug). Do not enable too many threads at one time, take up memory resources, you can avoid memory jitter. LeakCanary can detect memory leaks and release leaked resources.

Beijing-Zhang Xin-Front end: new technologies emerge one after another. What should the front end learn now?

51CTO: junior developers or learn HTML, CSS, JS, advanced developers, you have to consider some frameworks, such as Vue, React, and so on.

Beijing-Zhang Xin-Front end: when a new product comes out of the design drawing, how should a project structure be made to facilitate later maintenance?

51CTO: this requires multiple aspects, such as product manager, front-end development, background development, each in a different way. At the front end, first of all, you need to know what is public and what is not, which may start to be architected when there are no blueprints.

Do you still have any questions about these solutions? Welcome to join the 51CTO developer QQ Exchange Group 370892523 for discussion.

Advance notice of the next phase of the event: January 5

Keywords: distributed front-end APP

Activity Review: kicked Tribe Phase I: Spark offline Analysis Dimension

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.