TIBCO® Data Science

La data science est un sport d'équipe. Data scientists, citizen data scientists, utilisateurs métier et développeurs ont besoin d'outils flexibles et extensibles qui favorisent la collaboration, l'automatisation et la réutilisation des workflows d'analyse. Mais les algorithmes ne sont qu’une pièce du puzzle de l’analytique décisionnel avancée. Pour fournir des informations prédictives, les entreprises doivent davantage mettre l'accent sur le déploiement, la gestion et la supervision de modèles analytiques. Les entités opérationnelles performantes s'appuient sur des plateformes qui prennent en charge le cycle d'analyse de bout en bout tout en assurant sécurité et gouvernance. TIBCO® Data Science aide les entreprises à innover et à résoudre plus rapidement les problèmes complexes afin de transformer chaque prévision en résultat optimal.

Qu'est-ce que la data science ?

Data science is a multi-disciplinary approach to finding, extracting, and surfacing patterns in data through a fusion of analytics, domain expertise, and technology. Data science includes the fields of artificial intelligence, data mining, deep learning, forecasting, machine learning, optimization, predictive analytics, statistics, and text analytics.

TIBCO is a Leader in the Magic Quadrant for Data Science and Machine Learning Platforms 2019
2019 Gartner Magic Quadrant for Data Science and Machine Learning Platforms. Learn More

Bénéfices

Simplifiez l'analyse Big Data

Rely on huge scalability and broad support for data types, including unstructured data. A web-based interface with out-of-the-box, code-free operators eliminates the need for complex coding—and by performing all data prep and analytic computations natively within Hadoop and databases on-site or in the cloud, you both reduce the need to move data and improve performance.

La collaboration pour tous

Leverage the creativity of the entire cross-functional team, not just a few data scientists. The familiar web-browser interface makes it easy to share and mark-up, but lets you maintain control, security, version control, and auditability. Spotfire Data Science is the best solution for bringing machine learning, data, and people together to create operational solutions for the business

Reuse Data Science and Extend Your Power

Faster Visual Insights from Spark and Hadoop

Visual analytics are critical. But data prep leading up to them can be challenging. You don’t need scarce data scientists for this step. Go from data to a self-service visual interface with drag-and-drop operators to prep, blend, shape, and analyze data. Then display in a Spotfire dashboard where insights jump off the page.

From Data to Business Insight

Nous proposons une suite complète de fonctionnalités qui vont vous permettre d'atteindre les objectifs actuels et futurs de votre entreprise. L'intégration de la data science avec d'autres produits TIBCO pour créer des solutions métier : commencez à analyser dans Spotfire Data Science puis exportez vers Spotfire pour la visualisation. Utilisez TIBCO StreamBase® pour l'analyse en streaming, TIBCO Flogo® pour l'analyse dans l'edge et TIBCO Cloud™ LiveApps pour la création en libre-service d'applications métier.

Innovez en toute simplicité

Créez des modèles innovants à l’aide du deep learning, de l'intelligence artificielle prédictive et prescriptive et des méthodes d'analyse tout en encourageant une culture de l'analyse décisionnelle pour tous. Prototypez rapidement de nouvelles techniques avec Python, R, Scala et d'analyse intégrée, gages de flexibilité et d’exploration rapide de modèles. Décelez de nouvelles opportunités métier, créez des sources de revenus et monétisez des actifs pour en dégager un avantage concurrentiel durable.

Customer: University of Iowa Hospitals and Clinics transforms patient care with predictive analytics

Des données aux connaissances en un éclair

Customer: Tipping Point

Fonctionnalités

Project Stages & Roles

Five stages—define, transform, model, deploy, and act—organizes business stakeholders, data scientists, and engineers into teams that collaborate to tackle the problem at hand. Team members share and annotate data, scripts, and trained models within the project.

Workflow & Notebook Editors

Quickly build and share drag-and-drop machine learning and ETL workflows with the Workflow Editor. Easily run Python against Hadoop and RDBMS with the Notebook Editor. Notebooks, managed and versioned just like everything else in Chorus, make collaboration easy.

Anywhere Deployment

Deploy in AWS and perform data science at scale in Amazon EMR and Redshift. Run workflows on a regular schedule, and push PMML or PFA real-time scoring models to Cloud Foundry, AWS, or Google App Engine at the click of a button. Sync data to and from S3 with RedShift or MySQL.

Predictive Applications

Turn sophisticated data science projects into consumable UIs for employees and customers. Apply insights from scoring engines to the point of action. Use Touchpoints to embed findings into applications or bring them to mobile using the Touchpoints SDK.

Enterprise Security & Governance

Deploying analytics solutions against Kerberos, Ranger, and Sentry can be challenging; many open-source implementations aren’t suited. Chorus interacts with secured clusters, enabling advanced analytics on Hive and Spark to operate within IT data security policies.

In-Hadoop Computations

Reduce the need to move data and improve performance with analytic computations that run inside Hadoop and databases (Oracle, Pivotal, SAP HANA, SQL Server, others).

Demonstrations

Machine learning automatisé

Davantage de citizen data scientists aux commandes

Data science et IoT

Machine learning et analytique avancée

What's New

Request a demo today and see the latest new features including,

Rebranding from Alpine to Spotfire Data Science to fit the TIBCO user experience
Automatically connect your data science results to Spotfire with in-memory or in-source analyses
Configure Spotfire Data Science to connect to TERR for custom code within a visual workflow

Références

We applied predictive analytics to identify patients at risk of surgical site infection and changed our wound management strategies with remarkable improvements and outcomes for patients.