Building a FAQ Chatbot in Python – The Future of Information Searching 用Rasa NLU构建自己的中文NLU系统 MITIE and Chinese support Chinese_models_for_SpaCy Tuning Your NLU Model Rasa 入坑指南一：初识 Rasa rasa_chatbot_cn diet_classifier Rasa入门教程专栏 rasa文章导

Intent matching Matching algorithms two algorithms to match intents: rule-based grammar matching and ML matching. simultaneously attempts both algorithms and chooses the best result. Algorithm Pros Cons Rule-based grammar matching Accurate with a small or large number of training phrase examples.Models are updated quickly. Does not support the automated expansion entity option. ML matching Accurate with a large number of training phrase examples. Matching is fast. Inaccurate with a small

This tutorial shows you how to analyze and evaluate the quality of the training phrases supplied to your Dialogflow agent’s intents. The purpose of this analysis is to avoid confusing the agent with phrases irrelevant to the intents supplied to, or more relevant to, other intents. The approach you take is to generate semantic embeddings of the training phrases by using the TensorFlow Hub (tf.Hub) Universal Sentence Encoder module. You then compute cohesion and separation measurements based on the similarity between embeddings within the same intents and different intents.

Reference Joint_Intent_and_Slot_Classification.ipynb

Reference Named Entity Recognition using Spacy and Tensorflow

Background theory for recommendations the background theory for matrix factorization-based collaborative filtering as applied to recommendation systems. Collaborative filtering for recommendation systems Collaborative filtering relies only on observed user behavior to make recommendations—no profile data or content access is necessary. The basic assumption is that similar user behavior reflects similar fundamental preferences, allowing

Reference Delta Lake What is a Lakehouse?

Reference edX Insights edX Analytics Pipeline edx-analytics-pipeline luigi

This solution describes and compares the different design approaches for calling a machine learning model during a Dataflow pipeline, and examines the tradeoffs involved in choosing one approach or another. We present the results of a series of experiments that we ran to explore different approaches and illustrate these tradeoffs, both in batch and stream processing pipelines. This solution is designed for people who integrate trained models into data processing pipelines, rather than for data scientists who want to build machine learning models.

Implementing Recommendations Steps Set up a project Import your product catalog You can add items to your Recommendations AI product catalog individually by using the import Files or API. Information of the products sold to customers. This includes the product title, description, in stock availability, pricing, and so on. Record user events User events track user actions such as clicking on a product, adding an item to a shopping cart, or purchasing an item, and so on.

nginx gateway flask gunicorn microservice elasticsearch * 6 redis * 2 mysql * 3 mongo * 1

https://www.6aiq.com/article/1588904027930?m=0&p=1 全面理解搜索 Query https://m.sohu.com/a/341401199_473283/ https://blog.csdn.net/qq_40027052/article/details/78579587

https://www.elastic.co/guide/en/elasticsearch/reference/current/indices-aliases.html#filtered https://www.elastic.co/guide/en/elasticsearch/reference/current/mapping-routing-field.html https://www.elastic.co/guide/en/elasticsearch/reference/current/indices-aliases.html#aliases-routing https://discuss.elastic.co/t/what-is-best-indexing-strategy-for-multitenant-data/4353 多租户 规模 https://www.elastic.co/guide/en/elasticsearch/guide/current/user-based.html https://www.elastic.co/guide/en/elasticsearch/guide/current/shared-index.html https://www.elastic.co/guide/en/elasticsearch/guide/current/faking-it.html https://www.elastic.co/guide/en/elasticsearch/guide/current/one-big-user.html https://www.elastic.co/blog/found-multi-tenancy https://techblog.bozho.net/elasticsearch-multitenancy-with-routing/

ICS UCI CSE 163 CA106B Search Engine NYU Search Engine Page Rank Overview - Building a Search Engine | Coursera CSC 575 Intelligent Information Retrieval

https://en.m.wikipedia.org/wiki/Learning_to_rank https://docs.aws.amazon.com/elasticsearch-service/latest/developerguide/learning-to-rank.html https://github.com/dmlc/xgboost/tree/master/demo/rank https://github.com/o19s/ https://github.com/o19s/es-tmdb https://github.com/o19s/elasticsearch-learning-to-rank https://github.com/o19s/elasticsearch-ltr-demo https://github.com/o19s/hello-ltr https://github.com/ehsangolshani/ranklib-demo https://github.com/andrew-sn/sltr_demo https://blog.csdn.net/ctrip_tech/article/details/103519193 https://www.infoq.cn/article/we-are-bringing-learning-to-rank-to-elasticsearch https://github.com/o19s/rated-ranking-evaluator https://github.com/o19s/relevant-search-book https://github.com/o19s/word2vec-experiments https://github.com/o19s/splainer https://github.com/o19s/splainer-search https://github.com/elastic/elasticsearch-dsl-py

BaseService init_service 各个应用的模型 字典加载 service 对外暴露接口的统一 SearchService(BaseService) do_es_search 词库加载方案 多租户，每个租户词库隔离，每个租户一个JiebaTokenizer，一个wor

https://github.com/django-haystack/django-haystack/blob/master/haystack/backends/elasticsearch_backend.py 1 2 3 4 5 6 self.conn.indices.create( index=self.index_name, body=self.DEFAULT_SETTINGS, ignore=400 ) self.conn.indices.put_mapping( index=self.index_name, doc_type="modelresult", body=current_mapping )

Removal of mapping types https://www.elastic.co/guide/en/elasticsearch/reference/current/removal-of-types.html

服务名称 服务UUID APPID 服务Owner 创建时间 修改时间 ABtest 应用特性： 智能纠错 意图识别 槽位填充 搜索 个性化排序 平台特性： 多租户 服务解耦 租户级意图与槽