谁有pk10两期计划网站

  • Favorite
  • Setup as front page
  • work email
WeChat public account
share it
share to:
share it
Using Spatiotemporal Big Data to Analyze Residents' Consumption Situation
Source: Big Data Department Time: 2019-11-29

In the era of big data, no matter in the time dimension or the space dimension, the data shows an explosive growth. All social activities and behaviors in people's daily life will be recorded as a set of data, and the seemingly chaotic data often contains social The law and truth of operation. This article takes the field of consumption as an example to explore the application scenarios and methods of big data in the analysis of residents' consumption situation .

I. Challenges to Traditional Consumption Situation Analysis

Accurately grasping the consumption situation is of practical significance to understand and resolve potential hidden problems such as economic development, cultural construction, and social life. Therefore, consumption has always been one of the research hotspots of scholars at home and abroad. However, with the advent of the digital economy era, the booming development of new business models and new models has led to a new stage in the analysis of consumer consumption. 2019年国务院政府工作报告》明确提出, 充分发挥消费的基础作用,稳定国内有效需求,为经济平稳运行提供有力支撑。 The " State Council Government Work Report 2019" clearly stated that it is necessary to give full play to the fundamental role of consumption, stabilize domestic effective demand, and provide strong support for the stable operation of the economy. 分析方法尚 存在 一些 问题 ,亟待寻求新思路、新方法。 However, facing the new situation and new requirements, there are still some problems with the traditional analysis methods , and it is urgent to find new ideas and methods.

Based on statistical data, it is difficult to meet the timeliness and accuracy of analysis and judgment. CPI)、物价指数、恩格尔系数等, 但绝大部分 按月度甚至年度更新, 在支撑消费形势研判时具有一定的滞后性,难以适应瞬息万变的消费形势 Although there are already some mature index products in the consumer field, such as the consumer price index ( CPI), price index, Engel coefficient, etc., most of them are updated monthly or even annually, which has a certain lag when supporting the research and judgment of the consumption situation. It is difficult to adapt to the rapidly changing consumption situation .

It is mainly based on macro analysis, and it is difficult to sink to the meso-micro level. Under the new pattern of high-quality life, the focus on consumption is shifting from “quantity” to “quality” and from “overall” to “individual”, which has also led to a change in the focus of monitoring the consumption situation. Taking the consumption structure as an example, policy makers are not only concerned about the overall change trend, but also which consumer products in a certain area are upgraded or downgraded. Traditional indexes are relatively weak in solving such problems.

The data source is relatively single and it is difficult to support decision-making supervision requirements. With the rapid development of the Internet platform economy, new consumption models and business formats are emerging almost every day. Consumer products are emerging endlessly, which poses great challenges to the supervision methods and monitoring dimensions. In the past, monitoring and analysis methods relying on a single data source were difficult to support.

With the advent of the era of big data, the means of data acquisition, storage capacity, and analysis methods in the consumer field have all jumped sharply, bringing opportunities to solve the above problems. The purpose of this article is to explore the use of big data to further improve the ability to monitor and analyze the consumption situation of residents, serve decision-making in the field of consumption, and help stabilize consumption growth.

2. Analysis of the "quaternary" model of residents' consumption situation

By investigating current research hotspots in the field of consumption, and starting from the demand for decision-making services, a "quaternary" model of residents' consumption situation analysis based on "scale-structure-environment-hotspots" is established, and a vertical linkage consumption situation analysis is formed with this model as the core The three-tier architecture system of data, models, and application services horizontally serves the perspective of multiple subjects such as governments, enterprises, and individuals.

Figure 1   四元 ”模型架构图 Analysis of Resident Consumption Situation

The research focus of each part of the model is as follows:

(I) Estimated consumption scale

Analyze the changes in consumption scale under different business formats and modes, and conduct research on consumption scale trends from multiple perspectives such as online -offline, urban-rural, and consumer goods categories. 数据 ,在不涉及企业商业机密的前提下,开展多源数据的模型化融合,测算整体消费规模,分析存在的规律和周期。 The main method is to use statistical data such as sales records of leading e-commerce and offline merchants, as well as POS machine credit card records , to carry out a model fusion of multi-source data without calculating the business secrets of the enterprise, to calculate the overall consumption scale. , Analyze the existing laws and cycles.

(B) analysis of consumption structure

Analyze the shortcomings of residential consumption supply, the balance between urban and rural consumption upgrading, and the consumption upgrade in the dual market by category and region. The main method is to build a multi-indicator system based on mature indexes such as the consumer consumer price index and price index, and introduce a variety of indicators such as national income level, disposable income, and education level to measure the dynamic changes in consumption structure, and track, analyze, and forecast changes .

(3) Monitoring the consumption environment

Monitor the market consumption environment, analyze people's satisfaction with the consumer market, point out problems in the current market environment and future improvement directions, and carry out horizontal comparative analysis for key cities. The main method is to comprehensively evaluate the current consumption environment by using data such as public opinion, complaints, and social credit, and using technologies such as complex network analysis and text sentiment analysis.

(D) Focus on consumer hotspots

热点事件、热门商圈、热门行业 (如养老特别是社区养老服务、婴幼儿照护服务、旅游)和重点 消费品(如汽车、家政) 等,开展对消费问题的专题分析与研判。 Thematic analysis and research on consumer issues are conducted in response to hot events, popular business districts, popular industries (such as elderly care, community care services, infant and child care services, and tourism) and key consumer products (such as automobiles and housekeeping) . The main method is to use multi-source public opinion data such as news, Weibo, WeChat, blogs, forums, etc., and use meta search engine technology to deeply explore consumer hot issues.

Third, the analysis of residents' consumption situation based on big data

电商消费订单数据 用户手机信令数据 POI位置数据 重点城市商圈数据 ,以及 互联网主要媒体、论坛、博客、微博等渠道中与消费直接相关 的舆情数据等 Based on the "quaternary" model of residents' consumption situation analysis , this paper builds a large-scale monitoring and analysis platform for consumption data in the field of consumption to aggregate multi-source and heterogeneous massive consumption data, mainly including mainstream e-commerce consumer order data , user mobile phone signaling data , and POI location Data , data on business districts in key cities , and public opinion data directly related to consumption in major Internet media, forums, blogs, Weibo and other channels . At the same time, an index analysis was used to make a preliminary analysis of China's residents' consumption situation in the first half of 2019.

(I) Online Consumer Price Fluctuation Index

The online consumer price volatility index is used to measure the average price fluctuation of a single online order for residents. Based on the order data of mainstream e-commerce providers, the monthly consumer price volatility index of various categories was calculated. The analysis found that among the online consumer goods categories of urban residents in the first half of this year, the average price of electronic and electrical goods orders fluctuated the most, followed by clothing, and consumer goods prices. relatively stable. By region, the average price of merchandise orders in the eastern region of China fluctuates greatly. The online consumer price fluctuations in rural areas are basically the same as those in urban areas, but the fluctuations are generally smaller than in urban areas.

(2) Residents' Consumption Upgrade Index

人们在细 分品类 更多地 购买较贵 商品 ,消费升级指数可用于描述一段时间内居民消费升级情况。 Consumption upgrade refers to the fact that people buy more expensive goods in sub- categories . The consumption upgrade index can be used to describe the consumption upgrade of residents over a period of time. 9 1月 以来 ,消费升级指数呈明显上升趋势 2019 4月达到最大值 The analysis found that: Since January 1999 , the consumption upgrade index has shown a clear upward trend and reached its maximum in April 2019 . In terms of product categories, transportation and communications, other supplies and services, food, tobacco, clothing, and clothing consumption were the most obvious upgrades; daily necessities and services, education, culture, entertainment, and health care consumption were basically flat; residential consumption experienced 1 After the monthly rally, there was a continuous degradation.

(3) Consumption Heat Index of Offline Business District

Commercial district consumption heat index is used to measure the activity of residents' offline consumption. 80 个热门商圈在 2 019 年上半年的到访人流量(不含职住人群),热度前十商圈中,上海( 3个)、武汉(2个)、北京(2个)三城市占据七席,其中上海五角场的热度最高。 Using mobile phone mobile signaling data to analyze the traffic of visitors from 80 popular business districts in Beijing, Shanghai, Guangzhou, Shenzhen, Hangzhou, Wuhan, Chongqing, Chengdu, and Shenyang in the first half of 019 Crowd), among the top ten business districts, Shanghai ( 3), Wuhan (2), and Beijing (2) occupy seven seats, of which Shanghai Wujiaochang has the highest popularity.

(IV) Consumption Environment Assessment Index

By analyzing the information directly related to the consumer environment in major Internet media, forums, blogs, microblogs and other channels, and calculating the consumer environment assessment index, the results show that in the first half of this year, the national consumer environment ’s positive assessment accounted for 82.76%, a month-on-month increase of 1.73. Percentage points, a year-on-year decrease of 0.32 percentage points. The school's food safety protection operation and the centralized destruction of counterfeit and shoddy products were well received.

In addition, through the sound volume detection of hot topics and hot topics, it was revealed that the abuse of the "7-day no reason to return" event, the Xi'an Lee Star incident, and the African swine fever affected meat price rise incidents in the first half of 2019 are the most popular consumer hot events. Rights protection and protection of personal information are the consumer issues that netizens are most concerned about.

Fourth, the road to strengthen the consumption situation monitoring analysis of big data

This article proposes a "quaternary" model for analysis of residents' consumption situation , and uses multiple sources of spatiotemporal data to construct a consumption index to conduct a preliminary analysis of the consumption situation in the first half of 2019. An attempt by big data means in the field of consumption is an exploration of the path that big data supports economic decision-making. However, the construction of a consumption big data monitoring and analysis system cannot be separated from the participation of multiple subjects, creating a new pattern of post-event supervision of "government guidance, multiple subjects, and social participation". The following three major systems should be focused on in the early stages.

社会大数据归集汇聚体系。 Promote data collection and build a social big data collection and aggregation system. 电商企业、线下商超、零售企业 、大数据企业、行业协会商会和社会第三方机构,探索通过政府购买服务和合作机制建设 建立面向 线上消费平台、线下消费巨头 的业务数据采购和共享共用机制,依法依规推进 消费规模、消费结构、消费环境、消费热点 消费形势 相关的社会化数据资源按需、及时向相关监管部门开放。 For key e-commerce companies, offline supermarkets, retail enterprises , big data companies, industry associations, chambers of commerce, and social third-party organizations, explore the construction of government-purchased services and cooperation mechanisms to establish businesses for online consumer platforms and offline consumer giants Data procurement and sharing and sharing mechanism, according to laws and regulations to promote consumption scale, consumption structure, consumption environment, consumption hotspots and other consumption situation- related social data resources are opened to relevant regulatory authorities in a timely manner as needed.

Strengthen technical support and build a system for monitoring and analyzing consumption big data . 消费形势 相关的 研究 重点,搭建数据加工、建模环境,开展 数据融合、消费 模型训练、数据测试集校验建模, 构建消费领域传统统计指数与大数据指数融合模型, 形成分区域、分领域、分行业 消费大数据监测分析指标 体系。 Aiming at the research priorities related to the consumption situation such as consumption scale, consumption structure, consumption environment, consumption hotspots, etc. , set up a data processing and modeling environment, carry out data fusion, consumption model training, and data test set verification modeling, and build traditional statistical indexes in the consumption field Integrate the model with the big data index to form a consumption big data monitoring and analysis index system by region, sector, and industry .

Innovate supervision methods and build a support system for early warning and response to consumer risks. Based on big data analysis and mining algorithm models such as panoramic image analysis, subject feature identification, and risk rating in the consumer field, a consumer risk early-warning, response, and linkage treatment mechanism is constructed to provide national and local relevant business departments to conduct consumer situation monitoring analysis and consultation and consultation Technical support to effectively improve the efficiency of post-event supervision and response.

(Big Data Development Department: Wei Ying, Chen Dong, Huang Qianqian, Xing Yuguan)