拉杰夫古普塔

IBM美国公司在美国各地租赁了几处设施来运营其业务. The objective of this project was to improve facility utilization and reduce facility operations and lease costs, 还有许多业务约束.
我开发了Python整数编程算法来解决这个问题. 考虑业务约束使这个问题变得有趣和独特. I parameterized the optimization period (the period to look into the future) in the algorithm to provide multiple solutions. 客户特别喜欢这个特性.
技术:Python, plot，线性编程，包装纸浆

Newristics is a US-based global leader in applying decision-heuristic science to marketing. 使用启发式心理学(500多种不同的启发式)，它重写了每个营销信息.

I automated the message scorer process where a team compares the new message against the old one and analyzes it to rate how closely it depicts the heuristic.

然后使用文本清理对文本数据进行预处理, 文本归一化, 并生成归一化数据的一元图. I built two main models to solve this problem: XGBoost and deep neural network seq-to-seq learning.

对于XGBoost，我创建了大约900个特性(分为三个部分).
•NLP基本特征:信息的字数/比例/字符数, 单元/双单元的TF-IDF, TF-IDF相似性, 等等......
• Word embedding—similarity of self/pre-trained Word2vec/GloVe-weighted average embedding vectors (TF-IDF as weight), etc.
• Graph—degree of nodes, the intersection of neighbors, k-core/k-clique, degree of separation, etc.

I used the 深度学习 seq-to-seq model to enhance the sequence inference neural network architecture.

技术:Python, LSTM, gensim, GloVe, SpaCy, NLTK, Scikit-learn, TensorFlow, Keras, Jupyter笔记本, Git, 谷歌云平台

AbbVie公司. is a leading pharmaceutical company and introduced a drug whose market share slipped from 65% to 49%. 他们就三个主题进行了一项医生调查，以帮助制定战略计划.

We interviewed 119 physicians about HCV regiment attributes which impact the market driver, 55名医生关心病人的治疗, and 60 physicians about sales rep interaction and their impression about the message and interaction.
I worked closely with the C-level executive and product management team to analyze the survey and produced data/reports. This helped the product team and executive team to make more informed decisions—increasing market share through the identification of new opportunity, 目标细分市场, 并设计出巧妙的解决约束的新方法.
技术:Python, R, Plotly, Matplotlib, 回归, 集群, Association Rule

我参加了一个黑客马拉松来分类H&E染色乳腺癌组织学图像. 我们得到了一个最小的训练数据集(几百张图像). 增加分类器的鲁棒性, I used a strong data augmentation and deep convolutional feature extractor at different scales with pre-trained CNNs on ImageNet. 在这个特征集上，我应用了一个高度精确的梯度增强算法. I also avoided training neural networks on this amount of data to prevent suboptimal generalization.

技术:Python 3, Keras, NumPy, Pandas, SciPy, Scikit-learn

Problem: They have a large portfolio of products distributed to retailers through wholesalers (agencies). 有数千种独特的批发商- sku /产品组合.

In order to plan its production and distribution as well as help wholesalers with their planning, it is important for them to have an accurate estimate of demand at SKU level (34) for each wholesaler (60).

数据:使用60家机构、34家sku四年的数据进行预测.
•价格促销(美元/升):价格, sales, 以单位单位月为单位，按每百升的美元价值进行促销
•历史销量(百升):以代理商-库存-月为单位的销售数据
•天气(摄氏度):一个机构月份的平均最高温度
•行业苏打水销售额(百升):行业苏打水销售额
•事件日历:事件细节(体育、嘉年华等)
•行业量(百升):行业实际啤酒量
• Demographics: Demographic details (yearly income in dollars); used deep neural networks sequence to sequence learning for demand prediction

我开发了一个使用深度学习的卫星图像特征检测模型. 1KM × 1KM卫星图像有3波段和16波段两种格式. This multi-band imagery is taken from the multispectral (400-1040NM) and short-wave infrared (SWIR) (1195-2365NM) range.