Ruby Programmer • Developing Rails applications and in-house system management • Joined the company in July last year • Robot! → Me! → • In previous job, I worked as a research and development engineer at a manufacturer. (not a web developer) speakerdeck.com/hassaku
Pair programming ࡞ۀڥΛखܰʹγΣΞ ։ൃڥΛखܰʹೖख "QQFBM ࣗͷঢ়گΛݟ͑ΔԽ ϝϯόʔͷಘҙΛݟ͑ΔԽ アレコレと工夫した４つの例 Shared workspace Easily available development environment Visualized specialties of members Own status indicator Four examples of a little ingenuity
Downloading with Vagrant::Downloaders::HTTP... [vagrant] Downloading box: http://192.168.11.7:8080/packages/rails4_.box [vagrant] Extracting box... [vagrant] Verifying box... [vagrant] Cleaning up downloaded box... - Download vagrant project files % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 400 100 400 0 0 16162 0 --:--:-- --:--:-- --:--:-- 17391 x rails4_/ x rails4_/.vagrant x rails4_/Vagrantfile - Start VM [rails4_] Importing base box 'rails4_'... [rails4_] Progress: 100% [rails4_] Forwarding ports... [rails4_] -- 22 => 2222 (adapter 1) [rails4_] Creating shared folders metadata... [rails4_] Clearing any previously set network interfaces... [rails4_] Booting VM... [rails4_] Waiting for VM to boot. This can take a few minutes. [rails4_] VM booted and ready for use! [rails4_] Mounting shared folders... [rails4_] -- v-root: /vagrant - Login VM Welcome to Ubuntu 12.10 (GNU/Linux 3.5.0-17-generic x86_64) Last login: Sun Dec 2 05:10:04 2012 from 10.0.2.2 vagrant@quantal64:~$ εΫϦϓτ࣮ߦ 7.ͷμϯϩʔυ 7.ͷىಈ ϩάΠϯ ଈ࠲ʹ࡞ۀ։Մʂ Execute the script Download files Creating VM Login and start to work
ࣄܥ ಘҙ A log analysis of our communication tool shows specialties of members extracted from what they said automatically. Natural language processing Speciality Example of A’s profile Leafy used at the office daily is a communication tool like Twitter.
STQFD εϖοΫ άϦʔϯ ෳͷ͔Βੜ͞ΕΔ୯ޠͰߏ͞ΕΔจষू߹ͷ֬Ϟσϧ େྔͳσʔλͷػցֶशʹΑΓ ֤छύϥϝʔλΛਪఆ จষʹ͓͚ΔτϐοΫ֬ ɾɾɾ A generative model that explain why some words of the document are similar. / An example of a message τϐοΫ֬ʹج͖ͮ ੜ͞Ε֤ͨτϐοΫ ୯ޠग़ݱ֬ʹج͖ͮ ੜ͞Ε֤ͨ୯ޠ Generated words by probabilities Generated topics by probabilities Graphical model of LDA Probabilities of generating each topic in document Probabilities of generating each word in topic Each parameters are inferred from a massive collection of document.
ٕज़ܥͰཱͨͳͯ͘ ࣄܥͰಥग़͢Δϝϯόʔ Result Users ԣ࣠ɿ֤Ϣʔβ Correlation score between user profiles and coefficient of keyword x-axis y-axis Engineers with high scores on keywords about testing Keyword : “精算” A business unit member keeps higher score on nontechnical topic compared with technical topics.
QSPKFDU" QSPKFDU# QSPKFDU$ Especially, in case of company which has many projects lined up, such activity to help each other without barrier might lead to the 100% strength of an organization. organization