Ali big data team year-end summary template, your report how to make the boss satisfied?

The annual review gives you an idea of how much impact you made and what regrets you have left, and planning gives you another chance to make a fresh choice.

While planning is not as fast as change, especially in areas such as big data, we still need to be good at finding the sure parts of the uncertain in order to build a better foundation for next year’s work.

I. Review of work

In 20XX, the company implemented the requirements of intelligent operation, actively explored and excavated the value of big data, and promoted transformation and innovation. The current data asset scale exceeded XXPB (XX% year on year), label scale exceeded XX million (XX% year on year), XX data product system was further improved, and the annual completed development needs exceeded XX (XX% year on year). Monthly accurate marketing successful users exceeded XX million (XX% year on year), external value realization exceeded XX million yuan, big data internal and external value practice made continuous breakthroughs.

1. Data collection

Complete procurement, supply chain……. Such as more internal data collection, open up XX data collection chain, and constantly consolidate the foundation of big data. At present, the platform bears XXXX collection interfaces, XXX exchange interfaces and XXXXX tasks. The daily data collection volume reaches XXXT, increasing by XX% year-on-year.

2. Data modeling

Complete location, broadband, video…… And other models, in which the positioning accuracy reaches XX%, the label scale breaks through XX million (XX% year on year).

3. Data management

The XX project construction of big data management platform has been completed, and the development capabilities of Spark, HBase, IBM Stream, Gemfire, Sqlfire, Impla and other platforms have been added to provide end-to-end visual management capabilities from data dictionary, data development, machine learning, data auditing to operation and maintenance scheduling. The average data development period has been shortened from XX to XX, significantly lowering the threshold for front-line use.

Currently, the platform has entered XX teams, XX users, XX tenants, XX data scripts and XXXX big data model assets.

4. Data products

Completed the release and application of XX service interfaces such as user portrait, timing trigger, product recommendation, channel execution and effect evaluation, shortened the development cycle of marketing product application from XX months to XX days, supported XX innovative applications of big data marketing services throughout the year, and the monthly average number of successful marketing users exceeded XXX million.

4. Smart operation

Implement “XYZ Plan” of big data, carry out practical training of big data, and train XXXX people for XXXX times in total. Organized the big data modeling and application Competition, with the average number of participants exceeding XXX and the average number of participants exceeding XXX. Excellent models such as XXXX and XXXX that stood out in the competition significantly improved production;

The two flat channels “XXXX” and “YYYY” have been built, and the interaction between provinces and cities has been normalized. More than XX excellent models and applications have been promoted in the province. Currently, the number of users of the big data platform has increased from less than XX at the beginning of the year to XXX, and the number of self-developed models in the province has exceeded XXX, which has promoted the initial formation of the company’s big data application atmosphere.

Ii. Challenges

1, big data to drive the company’s operation is of great significance to wisdom, but the understanding of the current line for large data is relatively weak, skill control is insufficient, can’t apply big data effectively to day-to-day productivity, is an urgent need to deepen the big data the operation mode of the act in an opera in “plays”, promote the popularization and application of big data, improve operation and the breadth and depth of wisdom.

2. Ecological changes of big data business do not match business transformation. In the early stage, we failed to fully emancipate our minds and effectively respond to changes and transformation.

Plan for next year

Continue to strengthen the collection ability, strengthen the ability of big data exchange platform, realize heterogeneous cluster exchange, data subscription, database disaster, Flume and Kafka interface configuration, development and test tenant separation and other functions, improve the daily data processing capacity to XXXT, daily task scheduling capacity to XX million;

Relying on crawler engine, it comprehensively promoted the introduction of external Internet data, continuously expanded the analytic scope of content knowledge base, and the number of knowledge base exceeded XXXX million. Expand B/O/M domain data such as XX, increase the collection interface from XXXX to XXXX, and enterprise data convergence rate reaches XX%.

Build Agile Data Miner Delivery Cloud (ADMD) to automate the whole process of Data mining, including variable preparation, model training, model output, model verification and production deployment. Implement feature variable engineering, build data mining center, complete “five-in-one (R, Python, Spark, SPSS, TensorFLow)” high-low collocation machine learning environment integration, meet the needs of rapid modeling of various roles.

The key word of annual summary planning is value. If process work does not produce obvious value, try not to enter the summary report.