There are more than 50 APIs suitable for machine learning and prediction applications. What is your choice? (2018)-Alibaba Cloud Developer Community

What a wonderful thing it is to have ready-made modules or tools for people doing engineering projects and scientific research, you can complete the corresponding task without accessing the source code or understanding the details of the internal working mechanism. The common method is to call some APIs, namely some predefined functions, to provide applications and developers with the ability to access a set of routines based on a certain software or hardware. This article summarizes more than 50 APIs that are useful to machine learning industry, mainly in the following areas:

• Face and Image Recognition (Face and Image Recognition)
• Text Analysis, natural language processing, emotion Analysis (Text Analysis, NLP, Sentiment Analysis)
• Language Translation (Language Translation)
• Machine Learning and prediction (Machine Learning and prediction)

Face and image recognition

• 1.Animetrics Face Recognition: This API can be used to detect faces in images and match them with known Face datasets. In addition, you can add or remove a category from the searchable gallery, or add or remove a facial image from a category.
• 2.Betaface: This API provides face recognition and WEB service detection. Features include multi-face detection, face cropping, 123 face feature point detection, face verification and recognition, and similarity search in large databases.
• 4.Face imagery: provides facial recognition and detection services. You can call this API to train programs, such as Face detection, Face recognition, group Face recognition, Face creation, and information retrieval.
• 5.FaceMark: This API can detect 68 feature points on the front and 35 feature points on the facial contour.
• 6.FaceRect: This API is very powerful and free. It can detect faces or multiple faces in front and outline photos, and output the detection results in JSON format. In addition, the API can display the detected eyes, noses, facial features such as mouth.
• 7.Google Cloud Vision API: published on TensorFlow platform, enabling the model to learn and predict the content of the image. In addition, it can also help users search for their favorite images and obtain their annotations quickly and accurately. It can also be divided into thousands of categories according to ships, Lions, Eiffel Tower, etc., to detect facial related emotions and identify printed languages in multiple languages.
• 8.IBM Watson Visual Recognition: This API can understand the content of images, such as image markers, detect face, age and gender predictions, and also detect face similarity. You can also use this API to develop services based on your own tasks.
• 9.Imagga: This API can automatically assign labels to images so that images can be found. It is based on image recognition platform services.
• 10.Kairos: This platform allows developers to quickly add emotion analysis and face recognition to applications and services.
• 11.Microsoft Cognitive Service - Computer Vision: This API can analyze visual content in different ways based on input or user selection. For example, Mark images based on content, classify images, detect faces and return coordinates, identify content in specific areas, generate content descriptions, identify text in images, and Mark adult content.
• 12.Skybiometry Face Detection and Recognition: This API provides Face Detection and Recognition services. The latest version of this API can distinguish dark glasses from transparent glasses.

Text analysis, natural language processing, and emotional analysis

• 1.Bitext: provides the most accurate multilingual emotional theme analysis on the market. Currently, four semantic services are provided: Entity, concept extraction, emotion analysis, and text classification. The API supports eight languages. * 2.Diffbot Analyze: provides developers with tools that can identify, Analyze, and extract main content and chapters from any web page.
• 3.Free Natural Language Processing Service: provides Free services including emotion analysis, content extraction, Language detection, etc. This API is also very popular in the API market mashape.com.
• 4.Google Cloud Natural Language API: This API analyzes the structure and significance of text, including emotion analysis, entity recognition and text annotation.
• 5.Watson Natural Language Understanding: This API analyzes text to extract metadata from concepts, entities, keywords, categories, relationships, and semantic roles.
• 6.MeaningCloud Text Classification: This API performs pre-Classification tasks, such as Text extraction, Text tagging, and word segmentation.
• 7.Microsoft Cognitive Service - Text Analytics: This API detects emotions, key phrases, topics, and voices from Text. Other APIs in this group include Bing spelling check, language understanding, language analysis, and Web language models.
• 8.nlpTools: simple JSON for HTTP RESTful network services, used for natural language processing, and capable of emotional analysis and text classification for network news media.
• 9.Geneea: This API can analyze the original text provided, the text extracted from a given URL, or the documents directly provided.
• 10.Thomson Reuters Open Calais™: classify and link documents and entities (people, locations, organizations, etc.) based on natural language processing, machine learning, and other methods.
• 11.Yactraq influ2topicsis: This API converts audio-visual content into theme metadata through language recognition and natural language processing technologies.

Language Translation

• 1.Google Cloud Translation: This API can dynamically translate text between thousands of languages and allows the integration of websites and programs with this Translation service.
• 2.Google Cloud SPEECH-TO-TEXT: This API can apply powerful neural network models. Developers can convert audio into TEXT. This API supports 120 languages and their variants.
• 3.IBM Watson Language Translator: This API translates text from one Language to another, and also allows developers to develop custom models based on their own needs.
• 4.MotaWord: the manual fast translation platform supports 70 languages, and allows developers to query translation quotations, submit translation project documents and style guides, track project progress and obtain activity information in real time.
• 5.WritePath Translation: This API allows developers to integrate WritePath into other applications, providing word retrieval, submitting text Translation tasks, and obtaining appropriate information. * 6.Houndify: intelligent integration of voice and conversation into the product through an independent platform that always learns.
• 7.IBM Watson Conversation: build chat robots that understand natural languages and deploy them on message platforms and websites. Other APIs in this group can provide functions such as dialogue, Natural Language classifier, personality analysis, document conversion, and Tone Analyzer.
• 8.IBM Watson Speech: This API provides Speech-to-text and text-to-Speech conversion functions.

Machine learning and prediction

• 1.Amazon Machine Learning: query patterns from data. Typical API operations include fraud detection, demand prediction, target marketing, and click prediction.
• 2.BigML: provides cloud-hosted machine learning and data analysis services. You can create a data source and use a standard HTTP model to process standard supervised and unsupervised machine learning tasks.
• 3.Google Cloud Prediction: provides REST API to build machine learning models. These tools can help analyze data to add various features to applications.
• 4.co: provide product recommendation engines for e-commerce websites to accurately predict customers' intentions.
• 5.Hu: toma: helps developers around the world build and build commercial-level deep learning chatbot.
• 6.IBM Watson Retrieve and Rank: developers can load custom data into this service and use relevant algorithms to train machine learning models (Rank). The service output includes a series of related files and metadata.
• 7.indico: text analysis and image analysis are provided. This API is free of charge and does not require any training data.
• 8.Microsoft Azure Cognitive Service API: provide customers with personalized product suggestions and increase sales volume. Some new features are added, such as support for batch processing, better API resource manager, and simpler API interfaces.
• 9.Microsoft Azure Anomaly Detection API: detects abnormal event sequence data, such as memory leakage during memory usage.
• 10.Microsoft Cognitive Service-QnA Maker: extracts information into dialogues or easy-to-browse forms. Other APIs in this group provide services such as academic knowledge, entity links, and knowledge exploration.
• 11.Microsoft Cognitive Service - Speaker Recognition: enables applications to detect the current Speaker. Other APIs in this group provide services such as Bing language and custom identification.
• 12.MLJAR: provides services for prototype development and deployment of pattern recognition algorithms.
• 13.NuPIC: Numenta's Cortical learning algorithm based on Python and C ++, and maintained by the nupic community. This API allows developers to use the original algorithm to concatenate multiple regions and take advantage of the functions of other platforms.
• 14.PredicSis: This API has good insight into big data and improves marketing performance through predictive analysis.
• 15.PredictionIO: an open-source machine learning server built on Apache. Typical API methods include creating, managing, and recording users, retrieving items and content, and making personalized recommendations based on users.
• 16.RxNLP - Cluster Sentences and Short Texts: text mining and natural language processing server. Its statement clustering API can convert long text or Short text into logical groups.
• 17.Recombee: provides data mining, language query, and machine learning algorithm services through REST API. For more information about other APIs, see Mashape blog and Programmable Web website. If you have better API recommendations, please leave a message and reply. Thank you very much.

Author information

Pedro Lopez, personal homepage of data scientists: www.iampedrolopez.com this article is translated by Alibaba Cloud Yunqi community. The original title of the article is 50 + Useful Machine Learning & Prediction APIs, 2018 Edition, Translator: Begonia, Revision: Uncle_LLD. The article is a simple translation. For more details, please see the original text.