52 Useful machine learning and prediction APIs (various directional resources)

Author: Thuy T. Pham

Selected from the Heart of Kdnuggets Machine compilation participation: Wu Yu

Artificial intelligence is becoming the basic technology for a new generation of technology change, but developing artificial intelligence programs for their applications and businesses from scratch is expensive and often difficult to achieve the performance they want, but fortunately we have a large number of Ready-to-use APIs available to use. These APIs allow developers to integrate artificial intelligence services, such as intelligent identification, media monitoring and directional advertising, from other companies into their own products.

All the APIs in this article are divided into the following four groups:

Facial and image recognition

Text analysis, natural language processing, emotion analysis

Language translation

Prediction and other machine learning

Each group is sorted alphabetically, and the description comes from an introductory message from its Web site link on February 3, 2017.

Facial and image recognition

1.Animetrics Face recognition: This API can be used to detect facial characters on a picture and compare them to known facial features. The API also adds or removes search topics from a searchable gallery, or you can add or remove a face from a topic.

Link: http://api.animetrics.com

2.Betaface: A Web service for facial recognition and detection. Features include multiple face recognition, facial masking, 123 facial points (22 Basic, 101 advanced) detection, facial verification, identification, and large data concentration similarity search.

Link: HTTPS://WWW.BETAFACE.COM/WPA

3.Eyedea recognition: Focus on high-end computer vision solutions, focused on target detection and target recognition software. An identification service that provides eye, face, load, copyright and license plate detection. The greatest value of the API is its ability to instantly understand objects, users, and behaviors.

Link: http://www.eyedea.cz

4.face++: A facial recognition and detection service that provides detection, identification, and Analysis services that can be integrated into the application. Users can train the program by invoking the service to detect facial, facial, facial classification, create facial sets, create groupings, and get information.

Link: https://www.faceplusplus.com

5.FaceMark: The API provides the ability to identify 68 dots on the front face and 35 points on the side face photo.

Link: http://apicloud.me/apis/facemark/docs

6.FaceRect: This is a powerful API for facial detection and is completely free. The API can find the face (either positive or lateral) or multiple faces in a single photo, and give a JSON output for each face found. In addition, Facerect can find facial features (eyes, nose, and mouth) for each detected face.

Link: Http://apicloud.me/apis/facerect/demo

7.Google Cloud Vision API: The API is driven by powerful platforms such as TensorFlow, which enables you to learn and predict image content. It can quickly help users find their favorite pictures in a wide range of areas, and also has a rich annotation. It can be divided into thousands of categories (such as: Boats, lions, Eiffel Tower), to detect the face of the relevant expressions, but also to identify the picture of the printed text in many languages.

Link: https://cloud.google.com/vision

8.IBM Watson Visual Recognition: Able to understand the content of the image-the visual concept tag of the image, look for the human face, give the approximate age and gender, and look for a similar image in a set. You can also train the service by creating your own custom concepts.

Link: https://www.ibm.com/watson/developercloud/visual-recognition.html

9.Kairos: Allows users to quickly integrate emotional analysis and facial recognition into their applications and service platforms.

Link: Https://www.kairos.com/docs/api

10.Microsoft Cognitive service-computer Vision: This cloud-based API can analyze visual content in different ways based on input and user choice. For example, content-based tagging images, classified images, detecting human faces and returning coordinates, identifying specific areas of content, generating content descriptions, identifying text found in images, and labeling adult content.

Link: Https://www.microsoft.com/cognitive-services/en-us/computer-vision-api

11.Rekognition: Provides facial and scene recognition and optimization for social image applications. The rekognition API enables emotional recognition and gender detection using the features of the eyes, mouth, nose, and face, and can be used to determine gender, age, and mood.

Link: http://www.programmableweb.com/api/rekognition

12.Skybiometry face Detection and recognition: Provides facial detection and recognition services. The new version of the API contains features that distinguish sunglasses from transparent glasses.

Link: https://skybiometry.com/Documentation

Text analysis, natural language processing, emotion analysis

1.Bitext: Provides the most accurate multilingual theme-based market sentiment. Four semantic services are currently available: Entity and concept extraction, affective analysis, and text categorization. The API supports 8 languages.

Links: https://www.bitext.com/text-analysis-api-2/#How-accurate-is-the-analysis

2.Diffbot Analyze: Provides a developer tool that can be used to identify, analyze, and extract key content and chapters from any Web page.

Link: https://www.diffbot.com/dev/docs/analyze/

3.Free Natural Language Processing service: A free services including affective analysis, content extraction and language detection. This is a popular data API in the large-scale cloud API market.

Link: https://market.mashape.com/loudelement/free-natural-language-processing-service

4.Google Cloud Natural Language API: Analyze the structure and meaning of text, including affective analysis, entity recognition and text tagging.

Link: https://cloud.google.com/natural-language/reference/rest/

5.IBM Watson Alchemy Language: Can be used to teach computers how to read and perform textual analysis (e.g., to transform unstructured data into structured data, especially in the areas of social network monitoring, business intelligence, content recommendation, financial transactions, and directed advertising).

Link: http://www.alchemyapi.com/

6.MeaningCloud Text classification: This API can complete some of the predefined tasks, such as: Extract text, symbolic, remove disabled words, word-restore.

Link: https://www.meaningcloud.com/developer/text-classification

7.Microsoft Azure Text Analytics API: is a text Analysis Web service suite built with Azure Machine Learning. The API can be used to analyze unstructured text, which can be used for affective analysis, key phrase extraction, language detection, and subject detection. No training data is required.

Link: Https://docs.microsoft.com/en-us/azure/machine-learning/machine-learn

8.Microsoft Cognitive service-text Analytics: Detects emotions, key phrases, themes, and languages in the text.

Link: Https://www.microsoft.com/cognitive-services/en-us/text-analytics-api

Other APIs for this group (language cognitive services) include:

Bing Spell Check:https://www.microsoft.com/cognitive-services/en-us/bing-spell-check-api

Language Understanding:https://www.microsoft.com/cognitive-services/en-us/language-understanding-intelligent-service-luis

Linguistic analysis:https://www.microsoft.com/cognitive-services/en-us/linguistic-analysis-api

Web Language Model:https://www.microsoft.com/cognitive-services/en-us/web-language-model-api

9.nlpTools: is a simple JSON on HTTP RESTful Web services for natural language processing. It can decode the online news media for emotional analysis and text categorization.

Link: http://nlptools.atrilla.net/web/api.php

10.Semantic Biomedical Tagger: There is a built-in function to identify 133 types of biological entities, and can link them with existing knowledge base systems through semantic analysis.

Link: Http://docs.s4.ontotext.com/display/S4docs/Semantic+Biomedical+Tagger

11.Thomson Reuters Open calais™: The use of natural language processing, machine learning and other methods. Calais can classify and link your documents to entities (people, places, organizations, etc.), facts (person X for company Y) and events (Person Z is appointed chairman of Company Y in time X).

Link: http://www.opencalais.com/opencalais-api/