schedule Aug 8th 01:45 - 02:30 PM place Jupiter people 137 Interested

Context: The Search problem

GOJEK is a SuperApp: 19+ apps within an umbrella app. One of these is GO-FOOD, the first food delivery service in Indonesia and the largest food delivery service in Southeast Asia. There are over 300 thousand restaurants on the platform with a total of over 16 million dishes between them.

Over two-thirds of those who order food online using GO-FOOD do so by utilising text search. Search engines are so essential to our everyday digital experience that we don’t think twice when using them anymore. Search engines involve two primary tasks: retrieval of documents and ranking them in order of relevance. While improving that ranking is an extremely important part of improving the search experience, actually understanding that query helps give the searcher exactly what they’re looking for. This talk will show you what we are doing to make it easy for users to find what they want.

GO-FOOD uses the ElasticSearch stack with restaurant and dish indexes to search for what the user types. However, this results in only exact text matches and at most, fuzzy matches. We wanted to create a holistic search experience that not only personalised search results, but also retrieved restaurants and dishes that were more relevant to what the user was looking for. This is being done by not only taking advantage of ElasticSearch features, but also developing a Query semantics engine.

Query Understanding: What & Why

This is where Query Understanding comes into the picture: it’s about using NLP to correctly identify the search intent behind the query and return more relevant search results, it’s about the interpretation process even before the results are even retrieved and ranked. The semantic neighbours of the query itself become the focus of the search process: after all, if I don’t understand what you’re trying to ask for, how will I give you what you want?

In the duration of this talk, you will learn about how we are taking advantage of word embeddings to build a Query Understanding Engine that is holistically designed to make the customer’s experience as smooth as possible. I will go over the techniques we used to build each component of the engine, the data and algorithmic challenges we faced and how we solved each problem we came across.

 
 

Outline/Structure of the Case Study

  • Defining the context for the search problem
  • Why we need a Query Semantics Engine and how it can add value
  • Existing workflow and what was proposed
  • Inside the Query Semantics Engine: what the components are and how they fit into the picture
  • Building the components: two of the most important components of the query understanding workflow are Intent Classification and Query Expansion: in this talk I will focus on Query Expansion using word embeddings and enhancing the search results with the help of Intent Classification. I will also talk about Spell Correction as a preprocessing step.
    • Intent Classification

      Within GO-FOOD, people search for a variety of things - they search for cuisines (who wants to eat some Chinese?), dish names (let’s get some coffee), restaurant names (KFC!), ingredients (chicken dishes, please), meals types (such as dinner or breakfast), and even regions from where they want food (Hyderabadi biryani, anyone?). Identifying the broad category that the users is trying to find will greatly narrow down the possible results.

    • Query Expansion

      The purpose behind Query Expansion is to increase the recall of the search results. This is achieved by broadening the query to include additional words besides the original query. Since a majority of GO-FOOD users speak Bahasa (Indonesian) as their primary language, adding translations can greatly improve recall. Similarly, we also want to add words that exhibit semantic similarity. For example: if a user searches for “coffee”, we would add terms such as “cappuccino”, “mocha”, “latte” etc to the query.

  • How we brought all the components together when building the ElasticSearch Query
  • Overview of what kind of results were surfaced to the end user
  • Some parallel efforts we are working on, including Knowledge Graphs and Autosuggest.

Learning Outcome

  1. How to take advantage of word embeddings for building an intelligent search engine
  2. How to deal with data challenges such finding the right data and how to structure it for training
  3. How to choose the right metric when evaluating performance of a Search Engine

Target Audience

Data Scientists and Engineers looking to enhance their Search platform; Product Managers looking to do a cost-benefit analysis on whether it's worth investing in Search (hint: it is!); Other curious souls who want to learn about the applications of Word Embeddings!

Prerequisites for Attendees

An interest in the Search problem and a curiosity to find out what goes on behind the scenes.

A basic understanding of the following would be useful:
1. What word embeddings are and how the vector representations work
2. Building ElasticSearch queries using the DSL

schedule Submitted 7 months ago

Public Feedback

comment Suggest improvements to the Speaker
  • Dipanjan Sarkar
    By Dipanjan Sarkar  ~  6 months ago
    reply Reply

    Definitely an interesting topic, a few of questions:

    - When people order food usually they would want to be very specific some times with their query or very generic to see more options, what is the percentage of searches belonging to these categories\patterns or even any other potential search behavior

    - It would be good to briefly know at which stage in intent classification helping the search, would it be used more of a filter or something else?

    - You have mentioned query expansion which is pretty good, but considering coffee has so many variants, if you add them to QE, are you taking into account any past behavior to rank the order in which items are displayed or using some other factos too? because while embeddings are contextual they are pretty much very specific to the text content unless you are considering other aspects also in embeddings. Would be good if you can mention this in a few lines

    - Have you folks also considered other approaches in general like learning to rank etc in your search engine?

    • Ishita Mathur
      By Ishita Mathur  ~  6 months ago
      reply Reply

      Hi Dipanjan, those are definitely some very interesting questions.

      1) Users can represent various levels of intent, and this can vary from generic to very specific intents such as:

      1. Having an idea of the location, brand, cuisine or ingredient on the basis of which the user wants the create an order
      2. Knowing what dish to order but open to any restaurant, or knowing which restaurant to order from but being open to any dish
      3. Knowing which dish to order and which restaurant to order from

      Keeping this in mind, a majority of the user searches can be classified as either a dish, restaurant or an ingredient and the ratio of the search volume corresponding to these intents is approximately equal to 3:1:1 respectively.

      2) Intent classification is one of the first few steps, in the Query Understanding process, right after spell correction. We use the intent to narrow down the results based on cosine distance, and use that as a filter of sorts. In most cases, the class probability of a single class is fairly high, and we can use the specific intent to show results for that intent. However, there are cases where the intent is ambiguous, such as a query "burger", which can either refer to a burger dish, or to a restaurant Burger King. In these cases, we ask the user to choose with which intent he/she wrote the query, and then use the same filtering process as above.

      3) In Query expansion, we are using the overall past booking history to determine the expansion terms, as our objective is to move business metrics at this stage. We are evaluating personalising the expansion terms based on the user-specific history.

      4) Yes, we are using LTR to rerank our search results in order to add a personalisation factor to the search results. The LTR model uses features that are derived from customer order and view history, their interactions with different merchants, cuisines, and real-time features such as time of the day, ETA etc

      Hope this answers your questions! I'd be happy to provide clarifications and answer any follow up questions you might have.

      • Dipanjan Sarkar
        By Dipanjan Sarkar  ~  6 months ago
        reply Reply

        Wonderful, it was really good to see you go into this level of depth. Thanks for that.

        If you can do the same around these aspects during the talk and dive more around how you tackled these scenarios and built these components I think this can be a really good talk!

        • Ishita Mathur
          By Ishita Mathur  ~  6 months ago
          reply Reply

          Thanks, Dipanjan!

          I will definitely go into detail about our journey building the query semantics engine: both on what product decisions we made and why as well as on the workflow and type of data/models used in both intent classification and query expansion.

          I certainly look forward to communicating our learnings to a wide audience at ODSC.


  • Liked Dipanjan Sarkar
    keyboard_arrow_down

    Dipanjan Sarkar - Explainable Artificial Intelligence - Demystifying the Hype

    45 Mins
    Tutorial
    Intermediate

    The field of Artificial Intelligence powered by Machine Learning and Deep Learning has gone through some phenomenal changes over the last decade. Starting off as just a pure academic and research-oriented domain, we have seen widespread industry adoption across diverse domains including retail, technology, healthcare, science and many more. More than often, the standard toolbox of machine learning, statistical or deep learning models remain the same. New models do come into existence like Capsule Networks, but industry adoption of the same usually takes several years. Hence, in the industry, the main focus of data science or machine learning is more ‘applied’ rather than theoretical and effective application of these models on the right data to solve complex real-world problems is of paramount importance.

    A machine learning or deep learning model by itself consists of an algorithm which tries to learn latent patterns and relationships from data without hard-coding fixed rules. Hence, explaining how a model works to the business always poses its own set of challenges. There are some domains in the industry especially in the world of finance like insurance or banking where data scientists often end up having to use more traditional machine learning models (linear or tree-based). The reason being that model interpretability is very important for the business to explain each and every decision being taken by the model.However, this often leads to a sacrifice in performance. This is where complex models like ensembles and neural networks typically give us better and more accurate performance (since true relationships are rarely linear in nature).We, however, end up being unable to have proper interpretations for model decisions.

    To address and talk about these gaps, I will take a conceptual yet hands-on approach where we will explore some of these challenges in-depth about explainable artificial intelligence (XAI) and human interpretable machine learning and even showcase with some examples using state-of-the-art model interpretation frameworks in Python!

  • Liked Johnu George
    keyboard_arrow_down

    Johnu George / Ramdoot Kumar P - A Scalable Hyperparameter Optimization framework for ML workloads

    20 Mins
    Demonstration
    Intermediate

    In machine learning, hyperparameters are parameters that governs the training process itself. For example, learning rate, number of hidden layers, number of nodes per layer are typical hyperparameters for neural networks. Hyperparameter Tuning is the process of searching the best hyper parameters to initialize the learning algorithm, thus improving training performance.

    We present Katib, a scalable and general hyper parameter tuning framework based on Kubernetes which is ML framework agnostic (Tensorflow, Pytorch, MXNet, XGboost etc). You will learn about Katib in Kubeflow, an open source ML toolkit for Kubernetes, as we demonstrate the advantages of hyperparameter optimization by running a sample classification problem. In addition, as we dive into the implementation details, you will learn how to contribute as we expand this platform to include autoML tools.

  • Liked Sujoy Roychowdhury
    keyboard_arrow_down

    Sujoy Roychowdhury - Building Multimodal Deep learning recommendation Systems

    20 Mins
    Talk
    Intermediate

    Recommendation systems aid in consumer decision making processes
    like what to buy, which books to read or movies to watch.
    Recommendation systems are specially useful in e-commerce websites
    where a user has to navigate through several hundred items
    in order to get to what they’re looking for . The data on how users
    interact with the systems can be used to analyze user behaviour and
    make recommendations that are in line with users’ preferences of
    certain item attributes over others. Collaborative filtering has, until
    recently, been able to achieve personalization through user based
    and item based collaborative filtering techniques. Recent advances
    in the application of Deep Learning in research as well as industry
    has led people to apply these techniques in recommendation systems.
    Many recommendation systems use product features for recommendations.
    However textual features available on products are
    almost invariably incomplete in real-world datasets due to various
    process related issues. Additionally, product features even when
    available cannot describe completely a certain feature. These limit
    the success of such recommendation techniques. Deep learning
    systems can process multi-modal data like text, images, audio and
    thus is our choice in implementing multi-modal recommendation
    system.
    In this talk we show a real-world application of a fashion recommendation
    system. This is based on a multi-modal deep learning system which is able to address the problem of poor annotation in the product data. We evaluate different deep learning architectures
    to process multi-modal data and compare their effectiveness. We
    highlight the trade-offs seen in a real-world implementation and
    how these trade-offs affect the actual choice of the architecture.

  • Liked Subhasish Misra
    keyboard_arrow_down

    Subhasish Misra - Causal data science: Answering the crucial ‘why’ in your analysis.

    Subhasish Misra
    Subhasish Misra
    Staff Data Scientist
    Walmart Labs
    schedule 7 months ago
    Sold Out!
    45 Mins
    Talk
    Intermediate

    Causal questions are ubiquitous in data science. For e.g. questions such as, did changing a feature in a website lead to more traffic or if digital ad exposure led to incremental purchase are deeply rooted in causality.

    Randomized tests are considered to be the gold standard when it comes to getting to causal effects. However, experiments in many cases are unfeasible or unethical. In such cases one has to rely on observational (non-experimental) data to derive causal insights. The crucial difference between randomized experiments and observational data is that in the former, test subjects (e.g. customers) are randomly assigned a treatment (e.g. digital advertisement exposure). This helps curb the possibility that user response (e.g. clicking on a link in the ad and purchasing the product) across the two groups of treated and non-treated subjects is different owing to pre-existing differences in user characteristic (e.g. demographics, geo-location etc.). In essence, we can then attribute divergences observed post-treatment in key outcomes (e.g. purchase rate), as the causal impact of the treatment.

    This treatment assignment mechanism that makes causal attribution possible via randomization is absent though when using observational data. Thankfully, there are scientific (statistical and beyond) techniques available to ensure that we are able to circumvent this shortcoming and get to causal reads.

    The aim of this talk, will be to offer a practical overview of the above aspects of causal inference -which in turn as a discipline lies at the fascinating confluence of statistics, philosophy, computer science, psychology, economics, and medicine, among others. Topics include:

    • The fundamental tenets of causality and measuring causal effects.
    • Challenges involved in measuring causal effects in real world situations.
    • Distinguishing between randomized and observational approaches to measuring the same.
    • Provide an introduction to measuring causal effects using observational data using matching and its extension of propensity score based matching with a focus on the a) the intuition and statistics behind it b) Tips from the trenches, basis the speakers experience in these techniques and c) Practical limitations of such approaches
    • Walk through an example of how matching was applied to get to causal insights regarding effectiveness of a digital product for a major retailer.
    • Finally conclude with why understanding having a nuanced understanding of causality is all the more important in the big data era we are into.
  • Liked Venkata Pingali
    keyboard_arrow_down

    Venkata Pingali - Accelerating ML using Production Feature Engineering Platform

    Venkata Pingali
    Venkata Pingali
    Co-Founder & CEO
    Scribble Data
    schedule 8 months ago
    Sold Out!
    45 Mins
    Talk
    Intermediate

    Anecdotally only 2% of the models developed are productionized, i.e., used day to day to improve business outcomes. Part of the reason is the high cost and complexity of productionization of models. It is estimated to be anywhere from 40 to 80% of the overall work.

    In this talk, we will share Scribble Data’s insights into productionization of ML, and how to reduce the cost and complexity in organizations. It is based on the last two years of work at Scribble developing and deploying production ML Feature Engineering Platform, and study of platforms from major organizations such as Uber. This talk expands on a previous talk given in January.

    First, we discuss the complexity of production ML systems, and where time and effort goes. Second, we give an overview of feature engineering, which is an expensive ML task, and the associated challenges Third, we suggest an architecture for Production Feature Engineering platform. Last, we discuss how one could go about building one for your organization

  • Liked Joy Mustafi
    keyboard_arrow_down

    Joy Mustafi / Aditya Bhattacharya - Person Identification via Multi-Modal Interface with Combination of Speech and Image Data

    90 Mins
    Workshop
    Intermediate

    Multi-Modalities

    Having multiple modalities in a system gives more affordance to users and can contribute to a more robust system. Having more also allows for greater accessibility for users who work more effectively with certain modalities. Multiple modalities can be used as backup when certain forms of communication are not possible. This is especially true in the case of redundant modalities in which two or more modalities are used to communicate the same information. Certain combinations of modalities can add to the expression of a computer-human or human-computer interaction because the modalities each may be more effective at expressing one form or aspect of information than others. For example, MUST researchers are working on a personalized humanoid built and equipped with various types of input devices and sensors to allow them to receive information from humans, which are interchangeable and a standardized method of communication with the computer, affording practical adjustments to the user, providing a richer interaction depending on the context, and implementing robust system with features like; keyboard; pointing device; touchscreen; computer vision; speech recognition; motion, orientation etc.

    There are six types of cooperation between modalities, and they help define how a combination or fusion of modalities work together to convey information more effectively.

    • Equivalence: information is presented in multiple ways and can be interpreted as the same information
    • Specialization: when a specific kind of information is always processed through the same modality
    • Redundancy: multiple modalities process the same information
    • Complimentarity: multiple modalities take separate information and merge it
    • Transfer: a modality produces information that another modality consumes
    • Concurrency: multiple modalities take in separate information that is not merged

    Computer - Human Modalities

    Computers utilize a wide range of technologies to communicate and send information to humans:

    • Vision - computer graphics typically through a screen
    • Audition - various audio outputs

    Project Features

    Adaptive: They MUST learn as information changes, and as goals and requirements evolve. They MUST resolve ambiguity and tolerate unpredictability. They MUST be engineered to feed on dynamic data in real time.

    Interactive: They MUST interact easily with users so that those users can define their needs comfortably. They MUST interact with other processors, devices, services, as well as with people.

    Iterative and Stateful: They MUST aid in defining a problem by asking questions or finding additional source input if a problem statement is ambiguous or incomplete. They MUST remember previous interactions in a process and return information that is suitable for the specific application at that point in time.

    Contextual: They MUST understand, identify, and extract contextual elements such as meaning, syntax, time, location, appropriate domain, regulation, user profile, process, task and goal. They may draw on multiple sources of information, including both structured and unstructured digital information, as well as sensory inputs (visual, gestural, auditory, or sensor-provided).

    Project Demos

    Multi-Modal Interaction: https://www.youtube.com/watch?v=jQ8Gq2HWxiA

    Gesture Detection: https://www.youtube.com/watch?v=rDSuCnC8Ei0

    Speech Recognition: https://www.youtube.com/watch?v=AewM3TsjoBk

    Assignment (Hands-on Challenge for Attendees)

    Real-time multi-modal access control system for authorized access to work environment - All the key concepts and individual steps will be demonstrated and explained in this workshop, and the attendees need to customize the generic code or approach for this assignment or hands-on challenge.

  • Liked Ashay Tamhane
    keyboard_arrow_down

    Ashay Tamhane - Modeling Contextual Changes In User Behaviour In Fashion e-commerce

    Ashay Tamhane
    Ashay Tamhane
    Staff Data Scientist
    Swiggy
    schedule 7 months ago
    Sold Out!
    20 Mins
    Talk
    Intermediate

    Impulse purchases are quite frequent in fashion e-commerce; browse patterns indicate fluid context changes across diverse product types probably due to the lack of a well-defined need at the consumer’s end. Data from fashion e-commerce portal indicate that the final product a person ends-up purchasing is often very different from the initial product he/she started the session with. We refer to this characteristic as a ‘context change’. This feature of fashion e-commerce makes understanding and predicting user behaviour quite challenging. Our work attempts to model this characteristic so as to both detect and preempt context changes. Our approach employs a deep Gated Recurrent Unit (GRU) over clickstream data. We show that this model captures context changes better than other non-sequential baseline models.

  • Liked Venkatraman J
    keyboard_arrow_down

    Venkatraman J - Entity Co-occurence and Entity Reputation scoring from Unstructured data using Semantic Knowledge graph

    Venkatraman J
    Venkatraman J
    Sr. data Software engineer
    Metapack
    schedule 8 months ago
    Sold Out!
    20 Mins
    Talk
    Intermediate

    Knowledge representation has been a research for many years in AI world and its continuing further too. Once knowledge is represented, reasoning from that extracted knowledge is done by various inferencing techniques. Initial knowledge bases were built using rules from domain experts and different inferencing techniques like Fuzzy inference, Bayesian inference were applied to extract reasoning from those knowledge bases. Semantic networks is another form of knowledge representation which can represent structured data like WordNet, DBpedia which solves problems in a specific domain by storing entities and relations among entities using onotologies.

    Knowledge graph is another representation technique deeply researched in academia as well as used by businesses in production to augment search relevancy in information retrieval(Google knowledgegraph), improve recommender systems, semantic search applications and also Question answering problems.In this talk i will illustrate the benefits of semantic knowledge graph, how it differs from Semantic ontologies, different technologies involved in building knowledge graph, how i built one to analyse unstructured (twitter data) to discover hidden relationships from the twitter corpus. I will also show how Knowledge graph is data scientist's tool kit to discover hidden relationships and insights from unstructured data quickly.

    In this talk i will show the technology and architecture used to determine entity reputation and entity co-occurence using Knowledge graph.Scoring an entity for reputation is useful in many Natural language processing tasks and applications such as Recommender systems.

  • Liked Akash Tandon
    keyboard_arrow_down

    Akash Tandon - Traversing the graph computing and database ecosystem

    Akash Tandon
    Akash Tandon
    Data Engineer
    SocialCops
    schedule 7 months ago
    Sold Out!
    45 Mins
    Talk
    Intermediate

    Graphs have long held a special place in computer science’s history (and codebases). We're seeing the advent of a new wave of the information age; an age that is characterized by great emphasis on linked data. Hence, graph computing and databases have risen to prominence rapidly over the last few years. Be it enterprise knowledge graphs, fraud detection or graph-based social media analytics, there are a great number of potential applications.

    To reap the benefits of graph databases and computing, one needs to understand the basics as well as current technical landscape and offerings. Equally important is to understand if a graph-based approach suits your problem.
    These realizations are a result of my involvement in an effort to build an enterprise knowledge graph platform. I also believe that graph computing is more than a niche technology and has potential for organizations of varying scale.
    Now, I want to share my learning with you.

    This talk will touch upon the above points with the general premise being that data structured as graph(s) can lead to improved data workflows.
    During our journey, you will learn fundamentals of graph technology and witness a live demo using Neo4j, a popular property graph database. We will walk through a day in the life of data workers (engineers, scientists, analysts), the challenges that they face and how graph-based approaches result in elegant solutions.
    We'll end our journey with a peek into the current graph ecosystem and high-level concepts that need to be kept in mind while adopting an offering.

  • Liked Krishna Sangeeth
    keyboard_arrow_down

    Krishna Sangeeth - The last mile problem in ML

    Krishna Sangeeth
    Krishna Sangeeth
    Data Scientist
    Ericsson
    schedule 7 months ago
    Sold Out!
    45 Mins
    Talk
    Intermediate

    “We have built a machine learning model, What next?”

    There is quite a bit of journey that one needs to cover from building a model in Jupyter notebook to taking it to production.
    I would like to call it as the “last mile problem in ML” , this last mile could be a simple tread if we embrace some good ideas.

    This talk covers some of these opinionated ideas on how we can get around some of the pitfalls in deployment of ML models in production.

    We would go over the below questions in detail think about solutions for them.

    • How to fix the zombie models apocalypse, a state when nobody knows how the model was trained ?
    • In Science, experiments are found to be valid only if they are reproducible. Should this be the case in Datascience as well ?
    • Training the model in your local machine and waiting for an eternity to complete is no fun. What are some better ways of doing this ?
    • How do you package your machine learning code in a robust manner?
    • Does an ML project have the luxury of not following good Software Engineering principles?
  • Liked Pallavi Mudumby
    keyboard_arrow_down

    Pallavi Mudumby - B2B Recommender System using Semantic knowledge - Ontology

    45 Mins
    Case Study
    Intermediate

    In this era of big data , Recommender systems are becoming increasingly important for businesses because they can help companies offer personalized product recommendations to customers. There have been many acknowledged recognized successes of consumer-oriented recommender systems, particularly in e-commerce. However, when it comes to Business to-Business (B2B) market space, there has been less research and real-time application of such systems.

    In our case study, we present a hybrid approach of building a context-sensitive recommender system incorporating semantic knowledge in the form of domain ontology and a custom user- user collaborative filtering model in a B2B space. Using Engineering Products transaction data of an Instrumentation company, we demonstrate that this recommendation algorithm offers improved personalization, diversity and cold start performance compared to standard Collaborative Filtering based recommender system.

  • Liked AbdulMajedRaja
    keyboard_arrow_down

    AbdulMajedRaja - Become Language Agnostic by Combining the Power of R with Python using Reticulate

    AbdulMajedRaja
    AbdulMajedRaja
    Analyst (IC)
    Cisco Systems
    schedule 7 months ago
    Sold Out!
    45 Mins
    Tutorial
    Intermediate

    Language Wars have always been there for ages and it's got a new candidate with Data science booming - R vs Python. While the fans are fighting R vs Python, the creators (Hadley Wickham (Chief DS @ RStudio) and Wes McKinney (Creator of Pandas Project)) are working together as Ursa Labs team to create open source data science tools. A similar effort by RStudio has given birth to Reticulate (R Interface to Python) that helps programmers combine R and Python in the same code, session and project and create a new kind of super hero.

  • Liked Pushker Ravindra
    keyboard_arrow_down

    Pushker Ravindra - Data Science Best Practices for R and Python

    20 Mins
    Talk
    Intermediate

    How many times did you feel that you were not able to understand someone else’s code or sometimes not even your own? It’s mostly because of bad/no documentation and not following the best practices. Here I will be demonstrating some of the best practices in Data Science, for R and Python, the two most important programming languages in the world for Data Science, which would help in building sustainable data products.

    - Integrated Development Environment (RStudio, PyCharm)

    - Coding best practices (Google’s R Style Guide and Hadley’s Style Guide, PEP 8)

    - Linter (lintR, Pylint)

    - Documentation – Code (Roxygen2, reStructuredText), README/Instruction Manual (RMarkdown, Jupyter Notebook)

    - Unit testing (testthat, unittest)

    - Packaging

    - Version control (Git)

    These best practices reduce technical debt in long term significantly, foster more collaboration and promote building of more sustainable data products in any organization.

  • Lakshya
    Lakshya
    Applied Researcher-2
    Salesforce
    schedule 7 months ago
    Sold Out!
    45 Mins
    Talk
    Intermediate

    Deep learning has significantly improved state-of-the-art performance for natural language processing (NLP) tasks, but each one is typically studied in isolation. The Natural Language Decathlon (decaNLP) is a new benchmark for studying general NLP models that can perform a variety of complex, natural language tasks. By requiring a single system to perform ten disparate natural language tasks, decaNLP offers a unique setting for multitask, transfer, and continual learning. decaNLP is maintained by salesforce and is publicly available on github in order to use for tasks like Question Answering, Machine Translation, Summarization, Sentiment Analysis etc.

  • Liked Maulik Soneji
    keyboard_arrow_down

    Maulik Soneji / Jewel James - Using ML for Personalizing Food Search

    45 Mins
    Talk
    Beginner

    GoFood, the food delivery product of Gojek is one of the largest of its kind in the world. This talk summarizes the approaches considered and lessons learnt during the design and successful experimentation of a search system that uses ML to personalize the restaurant results based on the user’s food and taste preferences .

    We formulated the estimation of the relevance as a Learning To Rank ML problem which makes the task of performing the ML inference for a very large number of customer-merchant pairs the next hurdle.
    The talk will cover our learnings and findings for the following:
    a. Creating a Learning Model for Food Search
    b. Targetting experiments to a certain percentage of users
    c. Training the model from real time data
    d. Enriching Restaurant data with custom tags

    Our story should help the audience in making design decisions on the data pipelines and software architecture needed when using ML for relevance ranking in high throughput search systems.

  • Liked Kumar Nityan Suman
    keyboard_arrow_down

    Kumar Nityan Suman - Beating BERT at NER For E-Commerce Products

    Kumar Nityan Suman
    Kumar Nityan Suman
    Data Scientist
    Youplus Inc.
    schedule 10 months ago
    Sold Out!
    45 Mins
    Tutorial
    Intermediate

    Natural Language Processing is a messy and complicated affair but modern advanced techniques are offering increasingly impressive results. Embeddings are a modern machine learning technique that has taken the natural language processing world by storm.

    This hands-on tutorial will showcase the advantage of learning custom Word and Character Embeddings for natural language problems over pre-trained vectors like ELMo and BERT using a Named Entity Recognition case study over e-commerce data.