Keynote - Ron Galperin - Controller at City of LA
Keynote - Soheil Katal - Chief Information Officer at Los Angeles Unified School District
Keynote - Peter Loo - Acting Chief Information Officer, Los Angeles County
Keynote - Eva Pereira - Chief Data Officer at City of Los Angeles
Keynote - Karthik Ramasamy - Head of Streaming at Databricks
Keynote - Lakshmi Sharma - Chief Product and Strategy Officer at Fastly

From corporations to governments to private individuals, most of the AI community has recognized the growing need to incorporate ethics into the development and maintenance of AI models. Much of the current discussion, though, is meant for leaders and managers. This talk is directed to data scientists, data engineers, ML Ops specialists, and anyone else who is responsible for the hands-on, day-to-day of work building, productionalizing, and maintaining AI models. We'll give a short overview of the business case for why technical AI expertise is critical to developing an AI Ethics strategy. Then we'll discuss the technical problems that cause AI models to behave unethically, how to detect problems at all phases of model development, and the tools and techniques that are available to support technical teams in Ethical AI development.

- Enterprises are facing an imperative to grow their business to gain competitive advantage at breakneck speed. They need to achieve that by adding new value services efficiently and effectively. - To achieve growth from these new services, enterprises need new insights instantly from their constantly changing data. - Unfortunately, current data infrastructure solutions offer sub-optimal solutions that leave customers wrestling with to achieve their business goals. Why is real-time analytics so difficult? - Data freshness and fast responsiveness are both important and present technical challenges of their own. - User-facing analytics and operational analytics mean supporting thousands of users simultaneously. - You have to do a lot of unnecessary de-normalized tables (de-normalization jobs) in streaming pipelines. It is very difficult to build and maintain. - You can't easily update the data in realtime to analyze business changes. StarRocks re-invents real-time analytics with the only platform uniquely designed for the next generation real-time Enterprise, unleashing the power of business intelligence to help organizations accelerate their digital transformation. StarRocks makes real-time analytics a reality with the fastest, easy-to-use analytics platform on the planet.

The availability of AutoML (automated machine learning) with publicly accessible pre-trained models enable domain experts to automatically build high-quality custom ML applications without much requirement for ML model construction knowledge, which greatly speeds up the ML model development. AutoML has been an essential piece in the model-centric approach in the data science community. ' AutoDC (automated data-centric processing), similar to the purpose of AutoML, is a newly developed open source tool that enables domain experts to automatically and systematically improve datasets by fixing incorrect labels, adding examples that represent edge cases, and applying data augmentation, without much coding requirement and manual process. ' Coming these two frameworks enable the domain experts to improve both dataset and model concurrently and iteratively. ' In this talk, we will showcase 3 data science use cases and examples, which demonstrates the effectiveness of these two frameworks combined and how it empowers domain experts who don't know ML coding to do AI development.

The knowledge graph and graph database markets have long asked themselves: why aren't we larger? The vision of the semantic web was that many datasets could be cross-referenced between independent graph databases to map all knowledge on the web from myriad disparate datasets into one or more authoritative ontologies which could be accessed by writing SPARQL queries to work across knowledge graphs. The reality of dirty data made this vision impossible. Most time is spent cleaning data which isn't in the format you need to solve your business problems. Multiple datasets in different formats each have quirks. Deduplicate data using entity resolution is an unsolved problem for large graphs. Once you merge duplicate nodes and edges, you rarely have the edge types you need to make a problem easy to solve. It turns out the most likely type of edge in a knowledge graph that solves your problem easily is defined by the output of a Python program using the machine learning. For large graphs, this program needs to run on a horizontally scalable platform PySpark and extend rather than be isolated inside a graph databases. The quality of developer's experience is critical. In this talk I will review an approach to an Open Source Large Knowledge Graph Factory built on top of Spark that follows the ingest / build / refine / public / query model that open source big data is based upon.

Super Bowl Ad cost $7 million and each year a few Super Bowl ads go viral. The traditional A/B testing does not predict virality. Some highly shared ones reach over 60 million organic views, which can be more valuable than views on TV. Not only are these voluntary, but they are typically without distraction, and win viewer engagement in the form of likes, comments, or shares. A Super Bowl ad that wins 69 million views on YouTube (e.g., Alexa Mind Reader) costs less than 10 cents per quality view! However, the challenge is triggering virality. We developed a method to predict virality and engineer virality into Ads. 1. Prof. Gerard J. Tellis and co-authors recommended that advertisers use YouTube to tease, test, and tweak (TTT) their ads to maximize sharing and viewing. 2022 saw that maxim put into practice. 2. We developed viral Ads prediction using two scientific models: a. Prof. Gerard Tellis et al.'s model for viral prediction b. Deep Learning viral prediction using social media effect 3. The model was able to identify all the top 15 Viral Ads it performed better than the traditional agencies. 4. New proposed method is Tease, Test, Tweak, Target and Spots Ad.

Machine learning (especially deep learning) is becoming increasingly complex and expensive. Many companies build their core businesses (e.g., self-driving, credit card fraud detection, item recommendation, etc.) upon continuous model training and/or inferencing, which is typically performed with dozens or even hundreds of GPU machines on a (public or on-premise) cloud. While a cloud-based environment makes it possible for these jobs to dynamically scale with load changes (e.g., user requests), running these jobs under the cloud's pay-as-you-go pricing model incurs large monetary costs, which would rapidly grow with the model size/complexity, the size of datasets, and the number of users. BreezeML democratizes AI/ML by helping AI companies significantly increase their performance-per-dollar by making effective use of preemptible GPU instances. Rooted in years of research at UCLA and Princeton, BreezeML provides (1) a preemption-resilient software system that allows users to reliably run ML training/inference jobs on preemptible instances (such as spot instances) and (2) a virtual cloud interface that performs intelligent selection and scheduling of (spot and on-demand) instances to minimize the monetary costs with strong SLA guarantees. Currently, BreezeML provides two services: 1. An API server (http://windmill.breezeml.ai/apis/) that allows ML engineers to upload batch jobs for free trails. It also allows customers to use their own cloud (e.g., AWS) credential to log in and use BreezeML to run jobs under their own cloud configurations. 2. We provide a docker image of the Breeze runtime, which includes the Breeze-enhanced Pytorch/Tensorflow/XGBoost as well as a new K8S-based orchestration system that can be easily deployed in the user's local environment (compliant with the user's local security policies). Our runtime allows the user to (a) use cheap spot instances in the cloud or (b) sharing resources between (low-priority) training and (high-priority) inference jobs in their on-premise cluster, thereby significantly improving GPU resource utilization. Experiments across a wide range of vision, language, and classification models demonstrate that BreezeML improve the performance-per-dollar by an average of 3 times. Our approach also eliminates the need of resource over-provisioning in on-premise clusters by allowing (high-priority) inference jobs to safely preempt (low-priority) training jobs.

Transformer architecture was proposed by Google Brain in 2017 to process sequential data. Transformers can be used in Natural Language Processing (NLP) and Computer Vision applications. Transformer architecture is based on the concept of ‘Self-Attention’. Transformers replaced the RNN/LSTM architecture. The major advantages of Transformer architecture are that they are fast and bi-directional. The input text is fed into this architecture in parallel which allows faster processing. The leading Language models BERT (Bidirectional Encoder Representations from Transformers) and GPT (Generative Pre-trained Transformer), are built upon Transformer architecture. BERT was proposed by Google and GPT-1/2/3 was proposed by OpenAI. BERT Language Model is included in Google Search Engine. HuggingFace web portal provides many popular Transformers in different flavors. Transformer can be used for all Natural Language Processing (NLP) applications like sentiment analysis, translation, auto-completion, named entity recognition, automatic question- answering and many more. Transformers can also be used for generating artificial text, which is indistinguishable from text generated by humans. This talk will briefly cover the theory of Transformers. Next it will focus on how to fine tune the standard Transformer library (downloaded from Hugging Face portal) for a specific application.

Measure trends in a particular topic or search term across Google Search across the US down to the city-level. Integrate these data signals into analytic pipelines to drive product, retail, media (video, audio, digital content) recommendations tailored to your audience segment. We'll discuss how Google unique datasets can be used with Google Cloud smart analytic services to process, enrich and surface the most relevant product or content that matches the ever-changing interests of your local customer segment.

Business Intelligence is a technology-driven process that analyzes data and forms conclusions to help assist workers to make informed business decisions. From collecting to cleaning, to morphing, to displaying we will address the pain points, tips, and tricks on how to navigate this process of converting data from raw material to a final product. You'll learn: From a high level, the process of bringing data from the "back" to the "front". Tools and best practices for cleaning and displaying data. Understanding the foundations of business intelligence to better execute on objectives. The various ways of displaying data depends on circumstance.

Cancer is a complex disease which requires interactions between cell-intrinsic alterations and tumor microenvironment. The connection between epigenetics and genomic structure plays a key role in chromatin interactions and enhancer-promoter communications for transcriptional activities. Alterations of these components in oncogenic signaling pathway potentially cause cancer cell-intrinsic changes and inappropriate instructions to normal cell cycles, leading to abnormal cell growth. ' Topologically associating domains (TADs) and A/B compartments are the main structures of higher-order chromatin structure. These contact domains, chromatin states, super-enhancers, and histone modifications together regulate transcription and gene expression for normal/abnormal cell cycles. ' Several bioinformatics tools were utilized ' FANC for processing raw FASTQ data to Hi-C contact matrices, JuicerTools for obtaining the locations of contact domains on the entire genome, and CoolBox for visualizing chromatin contacts in different cell lines. ' High-resolution chromatin contacts showed dynamic interactions among chromosomal regions in different cell lines. ' Qualitative and quantitative features were comprehensively engineered from 3D chromatin folding and epigenetic regulators using available packages (scikit learn, pytorch, pandas, numpy, matplotlib, etc.). ' XGBoost multi-class classifier achieved the highest accuracy of 80.90% in classifying normal and cancer cell lines based on chromatin interactions, followed by Random Forest at 73.76% and TabNet classifier at 70.00%.

This talk describes an internal data product called Member Embeddings that facilitates modeling of member medical journeys with machine learning. Medical claims are the key data source we use to understand health journeys at Aetna. Claims are the data artifacts that result from our members' interactions with the healthcare system. Claims contain data like the amount the provider billed, the place of service, and provider specialty. The primary medical information in a claim is represented in codes that indicate the diagnoses, procedures, or drugs for which a member was billed. These codes give us a semi-structured view into the medical reason for each claim and so contain rich information about members' health journeys. However, since the codes themselves are categorical and high-dimensional (>10K cardinality), it's challenging to extract insight or predictive power directly from the raw codes on a claim. To transform claim codes into a more useful format for machine learning, we turned to the concept of embeddings. Word embeddings are widely used in natural language processing to provide numeric vector representations of individual words. We use a similar approach with our claims data. We treat each claim code as a word or token and use embedding algorithms to learn lower-dimensional vector representations that preserve the original high-dimensional semantic meaning. This process converts the categorical features into dense numeric representations. In our case, we use sequences of anonymized member claim diagnosis, procedure, and drug codes as training data. We tested a variety of algorithms to learn embeddings for each type of claim code. We found that the trained embeddings showed relationships between codes that were reasonable from the point of view of subject matter experts. In addition, using the embeddings to predict future healthcare-related events outperformed other basic features, making this tool an easy way to improve predictive model performance and save data scientist time.

Tech companies are making data products to empower us to work, explore, enjoy and create beyond what we can do now. With new Machine Learning based products paves new ways to capture, understand and communicate data. This drives us to question the tech product landscape. Namely, how these major tech companies have used the practices/capabilities from data science and usability mythologies from experience design to bring value to its users. Last year I spoke about how data and design can learn from each other, and this year, we'll reference today's real world examples to discuss how bridging data science with experience design influences data products. How companies are using design and data to introduce new products? How are methodologies in data science and experience design aiding in data product challenges? How might data collection shape product innovation? How might this affect privacy and security? Could evolving data products influence our behavior?

After interviewing companies like AirBnB, Stripe, Looker, and Revolut, Kirill learned that what differentiates the companies that have grown quickly and successfully is the quality of their data culture. Better data skill sets are not only nice to have, they offer a fundamental change within any company, and every metric is impacted positively when you can work with data. Kirill can share his five steps to implementing a strong data culture within any industry below: Aim for everyone to independently access data Don't rely solely on BI tools Level the playing field Automate what you can Bottom-up and top-down

1. Shopkick has over 30M downloads, but the userbase is very heterogeneous. Anecdotal evidence indicated a wide variety of users for whom the app holds long-term appeal. 2. Marketing and other teams challenged Analytics to get beyond basic summary statistics and develop a holistic segmentation of the userbase. 3. Shopkick's data science team used SQL and python to gather data, clean data, and then perform a data-driven segmentation using a k-means algorithm. 4. Interpreting the results is more work -- and more fun -- than running the algo itself. We'll discuss how we transform from ""segment 1"", ""segment 2"", etc. to something that non-analytics users (Marketing, Operations, etc.) could actually benefit from. 5. So what? How did team across Shopkick change their approach given what Analytics had discovered.

We have grown past the traditional reporting off of centralized EDW data store. Each department wants to be empowered to do their own analytics, but this creates new challenges in areas of governance, security, access and monitoring. How do we do this the right way? - Explore the concept of Data Mesh - Explore what is Data Product - Explore how to implement successful Self-Service - Pushing the limits of new capabilities

While great gender parity strides have been made in fields like web development and user experience (UX) design, data science "" relatively newer concentration"" still has a ways to go. Among technical education fields, data science is shown to have the lowest representation of female students, at just 35.3%. That lack of diversity is a serious issue. AI algorithms are susceptible to bias, so building them requires a team that includes a wide range of views and experiences. This session will focus on how to build an organization that supports the recruitment, mentorship, and career development of women in data.

As a bullied kid, I found refuge in computers and taught myself to code at 8. By 26, I had two successful tech exits and moved to NYC. A weekend party habit led to daily drug use and a spiral to heroin and homelessness. In 2016, after a friend�s overdose woke me up. I checked myself into rehab and quickly realized I was there for a bigger purpose. Healthcare is very broken. From legacy systems, inefficiencies, and poor customer experience. What if we could dramatically improve care models by leveraging data, personalizing treatment, and creating beautiful patient experiences? Ever worked in an industry that felt antiquated? Learn how we use MongoDB to transform addiction care and help people thrive in life!

The majority of traditional corporate valuation methods are solely based on tangible indicators ' sales growth, gross profit, cash flow, and operational performance ' which are the result and don't reflect the underlying process which makes the organization successful in the longer run. One of the most important intangible assets that don't appear directly on the balance sheet is human capital. In this research, we focus on the role that workforce skills, education, and knowledge play towards organizational success using state-of-the-art Machine Learning techniques. We find that investments in human capital not only play a crucial role in organizational growth but also have a causal relationship.

The frequency and severity of natural disasters are increasing. In response, governments, businesses, nonprofits, and international organizations are placing more emphasis on disaster preparedness and response. Many organizations are accelerating their efforts to make their data publicly available for others to use. Repositories such as the Registry of Open Data on AWS and Humanitarian Data Exchange contain troves of data available for use by developers, data scientists, and machine learning practitioners. In this session, see how a community of developers came together though the AWS Disaster Response hackathon to build models to support natural disaster preparedness and response.

The world is on a collision course between increasing water volatility and the need to produce an increasing amount of food. Water is the cornerstone of agricultural products for used directly for food and to sustain livestock. Indeed, water is the most important and highest cost input for agricultural production. To meet this reality, Pluvial.AI has developed a platform that uses a combination of remote sensing, meteorological data and analytics to provide farmers with near real-time assessments of plant health while determining the optimum watering needs to maximize crop yield.

Luxury yachts, football teams, and mansions are no longer safe havens for the illicit profits of Russian Oligarchs with ties to Putin. Assets are being identified and seized with benefits flowing to causes in Ukraine. This presentation covers: - How are friends and relatives of Putin sheltering immense profits - Graphs and other tools being used to identify sources & destinations of illicit wealth - Latest asset seizures - New regulations to expose hidden investors

* Develop an Analytic Center Framework (ACF) for Air Quality in support of the NASA AIST air quality technology innovation effort, Air Quality Analytics Center Framework (AQACF) * Harmonize air quality data sets, models, and algorithms to facilitate analysis and projections of air quality across those sources. * Demonstrate analysis application area will focus on air pollution in large cities (e.g., Los Angeles) * Generalize framework to facilitate analyses for air quality applications more broadly Coordinate with other AIST-18 efforts including GEOS-Chem, GCHP, Predicting What We Breathe, etc.

Getting customers engaged and excited about data architecture plans How to integrate UX practices into Data Engineering Data Governance is bullshit - why? Applying performance, scale and usability tests to your Data Engineering journey

The most modernized enterprises utilize polyglot architecture, applying the best-suited database technologies to each of their organization's particular use cases. To successfully implement such an architecture, though, you need a thorough knowledge of the expansive NoSQL data technologies now available. Attendees of this Data Con LA presentation will come away with: -- A solid understanding of the decision-making process that should go into vetting NoSQL technologies and how to plan out their data modernization initiatives and migrations. -- They will learn the types of functionality that best match the strengths of NoSQL key-value stores, graph databases, columnar databases, document-type databases, time-series databases, and more. -- Attendees will also understand how to navigate database technology licensing concerns, and to recognize the types of vendors they'll encounter across the NoSQL ecosystem. This includes sniffing out open-core vendors that may advertise as “open source,"" but are driven by a business model that hinges on achieving proprietary lock-in. -- Attendees will also learn to determine if vendors offer open-code solutions that apply restrictive licensing, or if they support true open source technologies like Hadoop, Cassandra, Kafka, OpenSearch, Redis, Spark, and many more that offer total portability and true freedom of use.

Data is the new oil. However, many organizations have fragmented data in siloed line of businesses. In this topic, we will focus on identifying the legacy patterns and their limitations and introducing the new patterns packed by Kafka's core design ideas. The goal is to tirelessly pursue better solutions for organizations to overcome the bottleneck in data pipelines and modernize the digital assets for ready to scale their businesses. In summary, we will walk through three uses cases, recommend Dos and Donts, Take aways for Data Engineers, Data Scientist, Data architect in developing forefront data oriented skills.

Data is all over the place, and what matters is how we manage that data and make sense out of it and take some meaningful data driven decision. In this session we will discuss about whole data engineering pipeline, starting from data collection, processing, analysis and visualization in a complete serverless fashion. We will pick some opensource dataset and shall store and process it on cloud(AWS). While the focus would be more on the general understanding of data pipeline aspects of data engineering, but during the process we will learn few of the AWS services which can help us to achieve our goal in an effective and efficient way.

MongoDB is now a Developer Data Platform. Come learn what�s new in the 6.0 release and Atlas following all the recent announcements made at MongoDB World 2022. Topics will include - Atlas Search which combines 3 systems into one (database, search engine, and sync mechanisms) letting you focus on your product's differentiation. - Atlas Data Federation to seamlessly query, transform, and aggregate data from one or more MongoDB Atlas databases, Atlas Data Lake and AWS S3 buckets - Queryable Encryption lets you run expressive queries on fully randomized encrypted data to meet the most stringent security requirements - Relational Migrator which analyzes your existing relational schemas and helps you design a new MongoDB schema. - And more!

It has been an exciting year in the open source database industry, with more choice, more cloud, and key changes in the industry. We will dive into the key developments over 2022, including the most important opensource database software releases in general, the significance of cloud-native solutions in a multi-vendor, multi-cloud world, the new criticality of security challenges, and the evolution of the open source software industry.

TurboTax is one of the well known consumer software brand which at its peak serves 385K+ concurrent users. In this session, We start with looking at how user behavioral data & tax domain events are captured in real time using the event bus and analyzed to drive real time personalization with various TurboTax data pipelines. We will also look at solutions performing analytics which make use of these events, with the help of Kafka, Apache Flink, Apache Beam, Spark, Amazon S3, Amazon EMR, Redshift, Athena and Amazon lambda functions. Finally, we look at how SageMaker is used to create the TurboTax model to predict if a customer is at risk or needs help.

in this session we discuss storing JSON in our database and how the tools in MySQL can make that task a bit easier. We will start with a discussion of why you might want to store data as JSON as opposed to other storage methods. This discussion will include some real world examples. We will then talk about the different ways in which MySQL can help us achieve that goal and how we might go about deciding what method is best. We will see examples of using the JSON data type to store data in a 'normal' table and use SQL commands to retrieve, filter and sort that data. We will also show how you can validate your JSON schema. Finally, we will talk about MySQL Document Store which allows developers to tap into the best features of a NOSQL database as well as a traditional SQL based database. Attendees of this session will have the following takeaways: - An understanding of when it might be a better alternative to store data as JSON - How to create a JSON column in a database and provide optional validation for the JSON schema. - How to store and retrieve JSON data, as well as filter and sort results based on data in our JSON object - How to set up MySQL Document Store and configure collections - How to store and retrieve documents, as well as filter and sort documents returned from MySQL Document Store

We will cover all aspects of optimizing your Snowflake Data Cloud including: *Dive deep into how Snowflake pay as you go costs work and how by utilizing our proven optimization tools - Snoptimizer SaaS Snowflake Optimizer - https://snoptimizer.com/ , scripts, and architecture techniques you typically can save 10-40++% on your existing Snowflake Account costs. *Explain how Snowflake Compute works and proven techniques on how to architect warehouses for both cost and performance efficiency. We cover in depth how snowflake scales BOTH out and in as well as up and down with compute resources. *Explain how Snowflake data storage works with Replication, Time-Travel, and Cloning. We explain these awesome features as well as their downsides if they are used and configured wrongly. *Cover Snowflake cloud services costs and features that have costs related to them, including Snowpipe, Search Optimization, Materialized Views, Auto-clustering, and other recent new cost based features that provide value at a cost. *Finally, we will discuss how you can ensure your Snowflake Account(s) are fully optimized not just for cost but also for security and performance on Snowflake. We will show you security and performance best practices as well as pitfalls to avoid.

For modern data teams, lineage is a critical component of the data pipeline root cause and impact analysis workflow, as well as a means of ensuring that data, models, and other data assets are healthy and reliable. That being said, the complexity of SQL queries can make it challenging to build lineage manually, particularly at the field level. Xuanzi Han, a member of Monte Carlo's data and product teams, tackled this challenge head-on by leveraging some of the most popular tools in the modern data stack, including dbt, Airflow, Snowflake, and ANother Tool for Language Recognition (ANTLR). In this talk, they share how they designed the data model, query parser, and larger database design for field-level lineage, highlighting learnings, wrong turns, and best practices developed along the way.

For modern data teams, lineage is a critical component of the data pipeline root cause and impact analysis workflow, as well as a means of ensuring that data, models, and other data assets are healthy and reliable. That being said, the complexity of SQL queries can make it challenging to build lineage manually, particularly at the field level. Xuanzi Han, a member of Monte Carlo's data and product teams, tackled this challenge head-on by leveraging some of the most popular tools in the modern data stack, including dbt, Airflow, Snowflake, and ANother Tool for Language Recognition (ANTLR). In this talk, they share how they designed the data model, query parser, and larger database design for field-level lineage, highlighting learnings, wrong turns, and best practices developed along the way.

As everyone’s data is growing, there is a growing need to become data-driven and harness decision-making from data. The existing and legacy approaches created silos, dark data, and risk. AWS Modern Data Architecture is alleviating these issues and enabling new capabilities through cost effective and purpose-built services and architecture. Learn how it works and how customers have used it to realize their business goals and outcomes.

CLTV is like a mythical creature in the world of business and data — everyone talks about it, some may even get a brief glimpse of it, but no one knows how to approach it, let alone dare to tame it. In this short but informative talk, you will be offered an opportunity to learn about the caveats and assumption of a metric that is on everyone's mind, while being exposed to seasoned frameworks and fresh perspectives. You will walk the well known road laid by the industry standard methods as well as unconventional paths hiding deep in the Data Science forests, with your trusted guide being Tatiana Tikhonova, a Lead Data Professional at Medallia.

Learn how UiPath UI+API capabilities enable end-to-end automation in the data journey, especially in the fragmented technology landscape with diverse data sources. UiPath accelerates data migration & testing activities, especially when it comes to data validation and data mapping activities—migrations & testing that are not easily feasible through ETL processes. Similarly, security professionals may currently use various AIOps tools to continuously improve the security posture of their organizations. But, those tools aren't enabling end-to-end security operations automation and alert fatigue continues to be a problem for Security Analysts. Through IT automation activities, UiPath provides easy-to-use, vendor-agnostic, robust security operations capabilities enabling full-scale security orchestration, automation, and response.

Data proliferation is growing as enterprises progress on their digital journey. Over 60% of an organization’s data is unknown, dormant, or underutilized. Imagine if you had the right data in the right context at the right time? What if all your data was connected, integrated, and truthful? Picture frictionless migrations, productive users and happy stakeholders. Boomi Enterprise Data Catalog and Preparation can take you there. In this hands on session, you will learn: - Learn why a Data Catalog is so important to a successful Data Governance strategy - See how to use Natural Language to Search and Discover data, using AI and ML to understand what data was cataloged - How data stewards can Collaborate, Contribute and Enrich the catalog by adding Context to the data - Develop Data Dictionaries and Business Glossaries to define data sets within the catalog - How it automatically Identifies and masks sensitive data - How ML uses the catalog to derive new data sets and load them directly into BI and Data Warehouse/Lake Tools to get the data science team the data they need more quickly

Overview of the CSU Data Architecture on moving on-prem ERP data to the AWS Cloud at scale using Delphix for Data Replication/Virtualization and AWS Data Migration Service (DMS) for data extracts

Most organizations considering open source and open core cloud technologies as part of their all-important data stack understand they need to rigorously evaluate the software's licensing terms and gauge the long-term health of its community and ecosystem. What still happens less frequently ' but is just as crucial to these risk assessments ' is developing a thorough understanding of the business models governing the commercial organizations attached to each data-layer technology being considered. You must discern the underlying motivations of the vendors or technology providers you depend on to deliver or support open source data-layer software (as well as those vendors with strong influence over its development and maintenance). By acutely understanding these incentives, you can identify if, where, and how they may map to possible risks to your enterprise's adoption and ongoing open source implementation. Don't limit the assessment to licenses and community health -- although both are still very key variables. This session will discuss specifics on what you need to look for and consider when vetting open source data technologies in the cloud as offered by: -- Businesses using OSS as the foundation of their own intellectual property -- Businesses that maintain total control offer the OSS they offer -- Major cloud providers

The session will include views from the panel (and myself) * Review the current challenges, volumes of events, staffing shortages, expertise deficiencies, siloed security controls, * Provide statistics from recent Ponemon Institute reports including the recent Cost of a Data Breach 2021 Report's findings in attack vectors, response/organizational impact and costs attributed to remote workforces, * Provide The impact in cost and response times of AI/Machine Learning etc. * Share the way's AI is used in law enforcement and critical infrastructure protection, * Discuss AI bias and evolving Trust and Validation requirements in AI systems, the necessity and value of AI insight to security and where the industry is moving in AI for security.

Join me to learn about a new end-to-end data testing approach designed for modern data pipelines that fills dangerous gaps left by traditional data management tools—one designed to handle structured and unstructured data from any source. You'll hear how you can use unique automation technology to reach up to 90 percent test coverage rates and deliver trustworthy analytical and operational data at scale. Several real world use cases from major banks/finance, insurance, health analytics, and Snowflake examples will be presented. Key Learning Objective 1. Data journeys are complex and you have to ensure integrity of the data end to end across this journey from source to end reporting for compliance 2. Data Management tools do not test data, they profile and monitor at best, and leave serious gaps in your data testing coverage 3. Automation with integration to DevOps and DataOps' CI/CD processes are key to solving this. 4. How this approach has impact in your vertical

A system for a single source of truth has been the holy grail of many Master data management solutions and is a critical pillar of many data governance models. Existing technology solutions and the myriad of tools fall short in many ways. The main challenge seems to be portions of the truth are embedded in applications, and the rules of the business govern the changes to this master data. The option of establishing a single source of truth outside the applications creates the challenge of managing and coordinating the change across many systems. Most rules to coordinate the change cannot be programmed into existing systems and have to be enforced manually. Existing tooling in this area provides connectors to application sources but cannot resolve conflicting changes to master data. Human intervention is needed. Most implementations of blockchain as implemented today are slow, cannot scale to support millions of transactions in a day and can be very slow and expensive. However, private Ethereum networks with custom “smart contracts” can potentially eliminate human intervention to resolve conflicts and improve change management. Furthermore, blockchain, with its immutable and distributed properties, can work across organizations as it can also serve as a data exchange mechanism. - Existing solutions for single source of truth have insurmountable challenges. - Existing Data Governance processes are manual, costly and error prone. - Immutability and distributed properties of blockchain make it a natural fit. - Smart Contracts can be leveraged to automate Data Governance processes. - Integrating Block chain with existing data governance process is the key.

Conversational AI is getting more and more widely used for customer support and employee support use-cases. In this session, I'm going to talk about how it can be extended for data analysis and data science use-cases ... i.e., how users can interact with a bot to ask analytical questions on data in relational databases. This allows users to explore complex datasets using a combination of text and voice questions, in natural language, and then get back results in a combination of natural language and visualizations. Furthermore, it allows collaborative exploration of data by a group of users in a channel in platforms like Microsoft Teams, Slack or Google Chat. For example, a group of users in a channel can ask questions to a bot in plain English like ""How many cases of Covid were there in the last 2 months by state and gender"" or ""Why did the number of deaths from Covid increase in May 2022"", and jointly look at the results that come back. This facilitates data awareness, data-driven collaboration and joint decision making among teams in enterprises and outside. In this talk, I'll describe how we can bring together various features including natural-language understanding, NL-to-SQL translation, dialog management, data story-telling, semantic modeling of data and augmented analytics to facilitate collaborate exploration of data using conversational AI.

Despite being the third most targeted industry for a cyberattack, manufacturing is one of the least prepared. Colonial Pipeline and meat plants were in the news but hundreds others are being attacked each day. Manufacturing is becoming increasingly digitised as the industry is adopting automation, IoT for monitor and control, to a greater extent than ever before. The Industrial Internet of Things (IIoT) is bringing artificial intelligence, cloud computing and robotics into factories. Cyber-physical systems can now integrate all aspects of the supply chain, including operational systems and information systems, and are taking the place of outdated, siloed machines. Any factory making use of these new technologies is known as a Smart Factory, and they're prompting what experts are calling the fourth industrial revolution, or Industry 4.0. Smart Factories will help the manufacturing industry considerably, as digital technology can offer greater efficiency in the production stage, better quality products with fewer mistakes, and more flexibility for working processes. So it's no wonder that manufacturers are moving quickly to update their factories by 2019, 75% of large manufacturers will have incorporated the Industrial IoT in their operations. And by 2022, the Industrial IoT market is expected to be worth $195.47 billion. I will discuss; * The Smart Factory & IoT / OT * Manufacturers are investing in digital * What cybersecurity risks do manufacturers face? * How can the manufacturing industry mitigate risk? * Three key questions every manufacturer should ask themselves before implementing IoT.

I believe that event-sourcing is the best way to implement persistence within a microservices architecture, but it hasn't always been the easiest solution to implement. In this talk, I will demonstrate how these two exciting technologies can be combined into one killer stack that simplifies event sourcing development. I will outline how to use DDD and CQRS concepts as a guide for developing an event sourcing food-delivery application based on Apache Pulsar and Quarkus that is 100% cloud native. Throughout this talk, I will demonstrate several different event sourcing design patterns across multiple microservices to feed multiple real-time dashboards that provide driver location tracking, and heatmaps. I will also highlight some patterns for using an event streaming platform as your event store.

We will address: (1) Community creation and engagement using tokens and NFTs (2) Organization of DAO structures and ways to incentivize Web3 communities (3) DeFi business models applied to Web3 ventures (4) Why Metaverse matters for new entertainment and community engagement models.

The California Cloud Workforce is an initiative in LA-area community colleges to develop the skills for future employment in Cloud and DevOps roles, spearheaded by Santa Monica College. Because there are more than 20 colleges participating and because the technology and required skills evolves rapidly, we have developed an NLP ensemble using federal data to identify occupational specializations in Cloud Computing, and the relevant coursework across many different institutions. Training and using the model consisted of several phases: * Extracting occupational data from the O*NET system and curricula data from Course Outlines of Record * Creating component models using DistilBERT, traditional NLP topic models, and the Bloom taxonomy of educational objectives * Ensembling the component models using PaCMAP * Deployment, and aggregating and visualizing results Using PaCMAP and DistilBERT produced a more parsimonious model that can leverage both transformers architecture and domain specific knowledge, and can be calibrated for Cloud Computing or other programs, and is so easy to manage that students deploy it themselves as part of their coursework.

The Manufacturing and Consumer Packaged Goods industries are still facing unprecedented negative pressures on their operating profits from ongoing supply chain disruptions. Many of these organizations have turned to AI to solve all their problems. Surprisingly (/sarcasm), many of these same organizations have failed to realize and sustain ROI from their investments in AI. In this discussion, we will review a few solutions to supply chain initiatives we've rolled out and highlight the keys to success that we've identified to help organizations sustain value from applications in AI. In particular, we will discuss end-to-end technical solutions in demand forecasting, machine learning for lead-time optimization, and full network optimization, as well as the governance and organizational structures we have to consider to make sure the value from these solutions scales throughout the organization.

Machine learning and artificial intelligence are becoming mainstays of the enterprise business world, but many entrepreneurs and small-business owners still shy away from investing in it. You might not think that small businesses need to use AI in the same way that large companies do, but that is a fallacy. AI can contribute to many facets of business success including data sorting, email marketing, and much more. In small companies, people think there is not enough data volume for AI applications and automatically dismiss it. Stop! AI can still be applied. The core thing is to integrate AI in a cohesive plan outlining the processes for collecting, processing, governing, and eliciting value from data, optimized for ML and data science. This hold true for small companies as well. How we frame our approach to AI initiatives will determine its success. Don't worry, I am not a zealot. I will not tell you AI and ML are the cure-all and will solve all your problems. Some tasks are particularly well suited to these techniques, but not all. What I love about them is the fact that they allow us to tackle difficult problems that might otherwise be too daunting.

1. What are feature stores. 2. Why are they important? 3. Using Databricks and the feature store offering to streamline ml. This hold true for small companies as well. How we frame our approach to AI initiatives will determine its success. Don't worry, I am not a zealot. I will not tell you AI and ML are the cure-all and will solve all your problems. Some tasks are particularly well suited to these techniques, but not all. What I love about them is the fact that they allow us to tackle difficult problems that might otherwise be too daunting.

Showcase how I pivoted from professional concert photography to data analytics at Blue Chip companies. I will show you the classes I took, the skills I developed, and how you can do the same in 2022 and 2023

Dashboards are fast to build and user-friendly due to great tools such as Tableau. But there are use cases where dashboards are a poor fit. A year ago the operations team at HopSkipDrive demanded a better solution to replace a suite of existing dashboards. Lacking suitable features in existing tools we built a system that sends real-time alerts to Slack channels. The initial project was a success. What happened over the next few months surprised us: users found 30 additional use cases for real-time alerts. Those alerts have become an integral part of the service improving safety and user experience on the platform. In addition, alert volume is used as a measure of defects, a component of our overall service level score. This talk will deconstruct what we learned. • When are dashboards a poor fit? • What use cases are well-served by real-time alerts? • How was the alert service built? • Using alert volume as a defect score.

utilize SCADA data and monitoring in real time Water utilities have real time SCADA data in all the water plants and facilities. It is challenging to get the large amount SCADA data and make use of them. * forecasting chemical concentrations in water in advance For water quality, water engineers should know how chemicals concentrations in water are changing. By knowing how the concentrations are changing, they can make right decisions on what to do to make water quality good by optimizing their operations of water quality. For example, by knowing chlorine, nitrite concentrations in advance they can decide when to flush the water pipelines again next. By doing flushing in right time just right amount, they can save costs and labors as well as efficiently operating water quality. * optimize asset spending allocations to spend using machine learning throughout given years. in regulated water utilities, budgets are decided every 3 years by investors. Given budgets and assets should be well allocated to spend so that the given budgets and assets are all well used when the 3 years end. It should not be spent too little or too much. By using machine learning, the optimal asset spending allocation is efficiently decided. Based on the information, project managers and general managers can spend right amount budgets in each month and hit the target in earning test when the 3 years end. * forecasting water supply is for predicting how much water should be purchased for shortage of water or drought. In Southern California, it is very important to forecast how much water will be supplied by producing the water because oftentimes, drought comes. If too little water is supplied, water utilities company purchase the shortage from different areas' water utilities companies. To set up the budget and purchase right amount of water, it is very important to forecast water supply by water production of the water utilities company. By applying machine learning, the water supply can be forecasted. * by forecasting water depth change in wells, engineers can know which wells will dry out by drought. By applying machine learning, the water depth of the wells are forecasted and the wells that will be dry are identified.

This talk focuses on our story of how we helped the government of California manage the emergency response to the COVDI-19 pandemic over the past few years. We'll go over how we became part of California's COVID-19 data task force and the challenges that were overcome to provide real-time and accurate COVID-19 statistics for California. Please see http://covid19.ca.gov for more information.

Data analysis presentation using Tidycensus on R platform to analyze trends on major race and ethnic demographics, data from U.S. Census decennial data.

Are you a frustrated developer who feels like they know enough? However, the success you thought you would have is out of reach. You see others who make better strides but why? Is it a skills gap? Come learn how to set yourself apart as a developer and learn the skills of influence and collaboration.

70% of AI leaders cannot explain how specific AI model decisions or predictions are made, and only 35% said their organization made an effort to use AI in a way that was transparent and accountable. Ethics and AI have become a central conversation in the tech industry, driven by the lack of understanding of data models, what information they are trained on, and the risk of bias. This is especially critical in sectors like healthcare, where algorithmic bias can leave out significant portions of a population and lead to devastating results. Attendees of this session will: - Understand the inherent challenges in utilizing artificial intelligence in a way that is transparent and accountable. - Why you need to pay more attention to your data models and how you are are using & training AI and deep learning. - How to develop a framework for identifying and overcoming inherent biases in data sets to ensure that your AI is driving more equitable products.

At Stitch Fix, data is integral to every facet of our business. We run a plethora of dataflows to transform raw data into features that models use to serve customers. We need to scale these dataflows both in code complexity, as we add new capabilities, and in data size, as we attain more customers. To ensure that these workflows don't devolve into an unmaintainable mess of spaghetti code (chalk full of in-place pandas operations), we built and open-sourced hamilton, a pluggable microframework to make scaling and managing complex dataflows easy. To use hamilton, one creates a dataflow by writing simple python functions in a declarative manner. The framework stitches them together, introducing an abstraction to configure and execute these dataflows. In this talk, we'll present the basic concepts of hamilton, discuss its impact at Stitch Fix, and share recent extensions to the project, including integrations with Dask, Spark, and Ray and why we're excited for its future.

OpenSearch is an incredibly powerful search engine and analytics suite for ingesting, searching, visualizing, and analyzing your data and it is fully open source. This Apache 2.0-licensed and community-driven collection of technologies harnesses an architecture that combines the powers of Elasticsearch 7.10.2, Kibana 7.10.2 and Apache Lucene. With OpenSearch, users gain a distributed framework featuring particularly powerful scalability, high availability, and database-like capabilities. Attendees at this DataCon LA presentation will come away understanding OpenSearch's architecture and its building-block technology components, including: -- Apache Lucene utilization. Learn how this high-performance Java-based search library utilizes Lucene's inverted search index to delivers incredibly fast search results (while supporting natural language, wildcard, fuzzy, and proximity searches). -- OpenSearch cluster architecture. An OpenSearch cluster is a distributed and horizontally-scalable collection of nodes, which are differentiated based on the operations they perform. Attendees will learn the specific functions of master, master-eligible, data, client, ingest nodes. -- Data organization. Understand how OpenSearch organizes data into indices (which contain documents, which contain fields). -- Internal data structures. Get an in-depth look at how OpenSearch achieves scalability and reliability by breaking up indices into shards and segments, and utilizes translogs. -- Aggregations. See how OpenSearch enables its advanced built-in analytics capabilities through the power of aggregations.

This presentation will discuss how breaches happen, the growth of these hacks and the cost incurred by various aspects of industry. These breaches continue to happen even with elaborate safeguards in place. The Banking/Financial Industry has sophisticated authentication systems, and yet the move toward Omni-Channel Banking remains elusive. It can more commonly referred to as broken. Knowing who you are connecting with is critical in this world...We have many methods - Biometrics, SMS, PKI Certificates, behavioral but mutual authentication is best by transferring the burden on the service and not the customer. We have Auto OTP: 1. Authentication to online service 2. User Authentication 3. User password auto updating Multi-Channel authentication has been fragmented in the past and moving to an Omni-Channel authentication provides security to the user that their PII data has not been compromised. The attendee will learn how to implement such a system that will provide the Zero Trust security so often needed throughout industry, education and Government.

Learn how to start building a thriving multicloud with Lyve Cloud. “I need a cloud solution that offers greater data protection and flexibility and that reduces TCO.” If this sounds familiar, you’re not alone. Many organizations today face the same demands, and they’re turning to multicloud as the solution. In this session, we’ll cover the benefits of such an approach and show how Lyve Cloud can integrate into your public cloud environment for a seamless multicloud experience.

Quantum computing is rapidly becoming commercially feasible. Many tech giants - Google, IBM, Honeywell, and Microsoft - are spending billions to far outpace Moore's Law. Last year achieved the major milestone of Quantum Supremacy where it was shown that a quantum computer could greatly outperform a classical computer. Quantum computing offers the promise of solving problems which would be impossible for a classical computer including optimization, anomaly detection, and material design. It also allows unhackable communication. In this presentation I will summarize what quantum computing is and why it is so important. I will sketch the landscape of the field including the hardware, software, and major customers at present. The tool most critical for data analysis - quantum machine learning - will be explained, along with the type of applications it is best suited for. Finally I will explain how you can take the first steps into leveraging quantum computing for your enterprise's benefit. -What is quantum computing -Who are the major players in the field -What is quantum machine learning and what types of problems can it address -How your company can take advantage of this

Digital identity is who you are in the digital world, and you should be able to control your own identity. In a world where Big Tech controls data on millions of its users, how do we claim our digital identities? Self-sovereign identity gives people the control over their identities by empowering them as intermediaries of identity-related transactions. People are the platform, not Big Tech or governments. Decentralized identity leverages technologies like blockchain and token-based economy to ensure that the identity infrastructure is not controlled by a few companies. Despite the decentralized infrastructure, federated identity enables identity information to be easily aggregated, processed, and available to use for good. Personalized identity individualizes the sharing of identity information (e.g. selective disclosure) based on different people's needs and sharing contexts. Secure identity makes sure that the access to identity information will be more secure, authenticated, and accountable. And lastly, intelligent identity helps streamline identity management.

Immersive media has been shown to be more effective for learning. It captivates the mind with interactions, involves the whole body through movement, requires eye-hand coordination, absorbs attention with learning by doing, and engages more of the senses. It’s a much stronger sensory experience, it's more memorable, and provides a greater opportunity to lead a participant through some form of transformation. Gamification becomes possible and reinforcements, both positive and negative, can be used to shape behavior. Concrete Oasis is a persistent, city-scale, augmented reality, civic exhibit intended for Pershing Square Park that addresses the water crisis with the intention of engaging the audience with interactions that will lead to lasting behavior change. Hear from the team designing this immersive exhibit about how they are using augmented reality to help shift the collective mindset towards a sustainable water future for LA.

Accessibility is about designing with usability for all people in mind. When we think about accessibility in data science most data scientists are familiar with using color-blind friendly colors in their data visualizations, but for many, thinking of examples and discussion beyond this is hard. We as data scientists need to design for more diverse populations to make a real impact on the world.

Top media and entertainment executives from major Studio, Streaming, Gaming, Advertising, and Social Media companies come together to discuss how they use structured datasets and their derivative insights to drive key business decisions that have forever changed how we consume entertainment content. These senior leaders will also share what data, expertise, and innovative thinking is needed both now and for many years on. Come see if you have some of the skills or desire to help the industry get to the future faster.

Startup Showcase Winners 1st Place - Aware Health 2nd Place - Griot 3rd Place - ServiceMob

Despite numerous proposals over the years, there is no one comprehensive federal law that governs data privacy in the U.S. There's a complex patchwork of sector-specific and medium-specific laws, including laws and regulations that address telecommunications, health information, credit information, financial institutions and marketing. The U.S. has hundreds of sectoral data privacy and data security laws among its states. U.S. state attorneys general oversee data privacy laws governing the collection, storage, safeguarding, disposal and use of personal data collected from their residents, especially regarding data breach notifications and the security of Social Security numbers. Some apply only to governmental entities, while others apply only to private entities, and some apply to both. The most comprehensive state data privacy legislation to date is the California Consumer Privacy Act (CCPA). Signed into law on June 28, 2018, it went into effect on January 1, 2020. The CCPA is cross-sector legislation that introduces important definitions and broad individual consumer rights and imposes substantial duties on entities or persons that collect personal information about or from a California resident.

Data Science tutorial is designed for people who are new to Data Science. This is a beginner level session so no prior coding or technical knowledge is required. Just bring your laptop with WiFi capability. The session starts with a review of what is data science, the amount of data we generate and how companies are using that data to get insight. We will pick a business use case, define the data science process, followed by hands-on lab using python and Jupyter notebook. During the hands-on portion we will work with pandas, numpy, matplotlib and sklearn modules and use a machine learning algorithm to approach the business use case.